UNIVERSITY OF CALIFORNIA 
 
 LIBRARY 
 
 OF THE 
 
 LOWER DIVISION 
 
 DIG 1 .3 19 .12..UBIER D1VIS1( 
 
 Afo. 
 
GIFT OF 
 Prof* Slate 
 
SCIENCE LECTURES AT SOUTH 
 KENSINGTON. 
 
 -s 
 
SCIENCE LECTURES 
 
 AT 
 
 SOUTH KENSINGTON. 
 
 BY 
 
 CAPTAIN ABNEY, R.E., F.R.S. 
 
 PROFESSOR STOKES. 
 
 PROFESSOR ALEX. B. W. KENNEDY, C.E. 
 
 F. J. BRAMWELL, ESQ., C.E., F.R.S. 
 
 PROFESSOR F. FORBES. 
 
 H. C. SORBY, F.R.S. 
 
 J. T. BOTTOMLEY, M.A., F.R.S.E. 
 
 SYDNEY H, VINES, B.A., B.Sc. 
 
 PROFESSOR CAREY FOSTER. 
 
 IN TWO VOLUMES. 
 VOL. I. 
 
 MACMILLAN AND CO. 
 
 1878. 
 
 [ The Riyht of Translation and Reproduction is Reserved.'] 
 
-r>TTX7 cj^mJLO 
 
 LONDON : 
 
 R. CLAY, SONS, AND TAYLOR, 
 BREAD STREET HILL, E.O 
 
WWER DIVISION 
 
 CONTENTS. 
 
 LECTUKE I. 
 
 BY CAPTAIN ABNEY, R.E., F.R.S. 
 
 PAGE 
 
 PHOTOGRAPHY .......,...,.. 1 
 
 V LECTURE II. 
 
 BY PROFESSOR STOKES. 
 
 THE ABSORPTION OF LIGHT AND THE COLOURS OF NATURAL 
 
 BODIES ......... ....... 33 
 
 FLUORESCENCE .... ........ . . . . 54 
 
 LECTUKE III, 
 
 BY PROFESSOR A. B. W. KENNEDY, C.E., OF UNIVERSITY COLLEGE, 
 LONDON. 
 
 THE KINEMATICS OF MACHINERY ......... 76 
 
 LECTURE IV. 
 
 BY F. J. BRAMWELL, ESQ., MEMBER OF THE INSTITUTE C.lf., F.R.S. 
 
 ON THE STEAM-ENGINE . . . \ 
 
 673230 
 
vi CONTENTS. 
 
 LECTUKE V. 
 
 BY PROFESSOR G. FORBES. 
 
 PAGE 
 
 RADIATION 173 
 
 LECTUKE VI, 
 
 BY H. C. SORBY, F.R.S. 
 
 MICROSCOPES 193 
 
 XECTURE VII. 
 
 BY J. T. BOTTOMLEY, M.A., F.R.S. B., DEMONSTRATOR OF NATURAL 
 PHILOSOPHY IN THE UNIVERSITY OF GLASGOW. 
 
 ELECTROMETERS 216 
 
 LECTURE VIII. 
 
 BY SYDNEY H. VINES, B.A., B.SC., FELLOW OF CHRIST'S COLLEGE, 
 CAMBRIDGE. 
 
 ON THE APPARATUS RELATING TO VEGETABLE PHYSIOLOGY 249 
 
 LECTURE IX. 
 
 BY PROFESSOR CAREY FOSTER. 
 
 ELECTRICAL MEASUREMENTS 264 
 
SCIENCE LECTURES AT SOUTH 
 KENSINGTON. 
 
PHOTOGRAPHY: ; 
 
 TWO LECTURES. 
 
 BY CAPT. ABNET, R.E., F.R.S. 
 
 LECTUEE I. 
 
 IN Fe"nelon's fables, under the title of Voyage Suppose, 1690, 
 a visit to the Isle of Wonders is described, and in that book 
 we read 
 
 " There was no painter in that country, but if anybody 
 wished to have the portrait of a friend, of a picture, a beautiful 
 landscape, or of any other object, water was placed in great 
 basins of gold or silver, and then the object desired to be 
 painted was placed in front of that water. After a while 
 the water froze and became a glass mirror, on which an in- 
 effaceable image remained." 
 
 Such was a fancy which, though then of a most improbable 
 nature, presented itself to the mind of a French king's tutor 
 some hundred years ago. In its broad aspect it became a 
 reality when the first Daguerrean image was obtained ; 
 though accomplishment was, in a measure, effected in the 
 early asphaltum prints of Niepce. It is with the realisation 
 of this dream that we have to deal this morning; with that 
 realisation which has furnished people, to be numbered by 
 thousands, with the means of subsistence, and has created 
 fortunes in some few instances ; and which, has put a new 
 power into the hands of men of science in their investi- 
 gations. 
 
 S> B 
 
2 LECTUEES TO SCIENCE TEACHERS. 
 
 We shall only cast a rapid glance over the early history of 
 photography, and endeavour to show as far as possible how 
 it was that the great advances in it have been made. 
 
 In 1777 Scheele, of Stralsund, in Sweden, was the first 
 who actually carried out investigations into the action of light 
 on silver chloride. Before his time it was well known that 
 " luna cornua " (as silver chloride was termed) blackened in 
 the presence of light, but he arrived at the fact that a chemi- 
 cal cbange wa^s brought about by the light. He found that 
 silver chloride blackened by this agency on being treated 
 \viVn ammonia yielded up metallic silver, whilst if the ex- 
 posure cook place beneath water a soluble substance was 
 separated, which, when silver nitrate was applied, gave 
 fresh silver chloride. This was an important investigation, 
 but no fruits resulted from it till many years later. 
 
 Wedgwood, in 1802, next called attention to photographic 
 action, in a paper read before the Eoyal Institution, entitled, 
 "An Account of a Method of Copying Paintings on Glass, 
 and of making Profiles by the Agency of Light upon Nitrate 
 of Silver," with Observations by H. Davy. This was the 
 first published account of producing photographs. 
 
 Wedgwood used white leather or white paper as a sub- 
 stratum (using the former in preference to the latter) on 
 which was brushed silver nitrate. In his paper he entered 
 into details of his process, and admitted that the images so 
 obtained could not be fixed or rendered permanent. Davy 
 compared Wedgwood's results when using silver nitrate, with 
 those obtained by silver chloride, and found that the latter 
 compound was more susceptible to darkening by the action 
 of light than the former ; but in neither case could he fix the 
 images. 
 
 The next eminent man who made essays on what we now 
 call photography (or sun-writing) was Nicephore de Mepce. 
 He commenced his experiments in 1814, and in 1827 
 wished to communicate an account of them to the Eoyal 
 Society of London ; his paper was not received, owing to the 
 details of the process being kept a secret. We now know 
 that his process was founded on the action that light pro- 
 duced on bitumen of Judaea (more commonly known, perhaps, 
 as asphaltum) ; he found tbat this body when exposed to light, 
 became insoluble in the usual menstrua. Thus, if a thin 
 coating were given to a metal plate and when dry exposed 
 
PHOTOGRAPHY. 3 
 
 in a camera to the action of light controlled by a lens, the 
 part acted strongly upon by light would become insoluble, 
 and thus when a solvent was applied (such as naphtha or 
 petroleum) the shadows would be represented by the metal 
 plate, whilst the lights would be formed of the dark resin. 
 The image in this case would be reversed in the character 
 of its shades, unless the black body could be whitened, 
 and the metal blackened. After many experiments with this 
 object in view, Niepce applied iodine to the image obtained 
 with the asphaltum. These bitumen pictures are still in 
 existence, one or more being in the British Museum at the 
 present time. 
 
 In 1824 Daguerre, who was devoted to painting, com- 
 menced similar experiments to those of Niepce, each working 
 in secrecy and unknown to the other. In 1826, however 
 through the want of reticence of an optician, who was a 
 acquaintance of both, the fact that each of them was workin 
 in the same direction was learnt by the other, and in 1829 
 they entered into a kind of partnership. Here it may be 
 that Daguerre first learnt the treatment of metallic plates 
 with iodine, and watched the action that took place in the 
 light, when silver was employed to receive the layer of 
 asphaltum. At any rate to Daguerre belongs the discovery 
 of the action of light on iodide of silver surfaces, and also 
 the merit of producing a picture in the camera with but a 
 short exposure. When I say short, I mean short compared 
 with that given to the bitumen plates, for with such it took 
 six or eight hours to obtain an image. Working with 
 silver plates, which had been subjected to the vapour of 
 iodine, he succeeded at first in obtaining visible images 
 with prolonged exposure ; but whilst endeavouring to obtain 
 them in a moderate time he waded through endless experi- 
 ments, and only chance befriended him at last. It thus 
 occurred : Having exposed some iodized plates in the camera 
 and obtained no results, he placed them away in a cup- 
 board containing a medley of chemicals. On opening it 
 some time afterwards to procure an old plate to clean for 
 fresh trial he found, much to his astonishment, one of them 
 with a fully developed image upon it. I will not exhaust 
 your patience by detailing how he traced the agency at work 
 which had caused this development. Suffice it to say that 
 it was found to be mercury (which vaporises at ordinary 
 
 B 2 
 
4 LECTURES TO SCIENCE TEACHERS. 
 
 temperatures) which had collected on the parts acted upon by 
 light. 
 
 In June 1839 Daguerre's discovery was announced, and 
 in August of the same year published to the world, and a 
 pension of 6,000 francs per annum given to him by the 
 French Government, whilst at the same time 4,000 francs 
 was allotted to Niepee the Younger, who had succeeded to 
 the partnership with Daguerre after the death of his uncle. 
 
 An outline of the daguerreotype process is as follows : 
 
 A copper plate is silvered by the electro-plating process or 
 any other convenient method, and after very careful cleaning, 
 the silver surface is exposed in the dark to the action of 
 iodine vapours. The iodine combines with the silver, and 
 the metallic surface becomes covered with silver iodide, first 
 canary coloured, then rose, then blue, and so on, the colour 
 being dependent on the thickness of the Layer of silver iodide 
 produced. When canary-coloured it is supposed that the 
 surface is in the best condition for receiving the impact of 
 light. 
 
 It will now be convenient to point out the chemical 
 change that really takes place in the ordinarily employed 
 silver salts when exposed to the action of light. Under 
 certain circumstances, when- subject to its impact, silver 
 iodide (which for our purpose we will call Ag 2 I 2 ) throws 
 off one atom of iodine, and we get subiodide of silver 
 (Ag 2 I), a slightly black body, left behind. Scheele proved 
 "by his experiments that silver chloride (Ag 2 C1 2 ), when acted 
 upon by light, gave off chlorine (Cl), and we now know 
 that the blackened product is sub-chloride of silver (Ag 2 Cl). 
 
 Similarly, silver bromide (Ag 2 Br 2 ) is converted into the 
 sub-bromide (Ag 2 Br). 
 
 Pure and dry silver chloride will change in the light. 
 Pure and dry silver bromide will also change in the light, 
 but not so readily as the chloride. Pure and dry silver 
 iodide is unaffected by light, unless any body which will take 
 up iodine be present ; even moisture will induce the change if 
 the impact of light be prolonged. 
 
 The sensitiveness of both the chloride and bromide is 
 materially increased by the presence of any body which will 
 absorb chlorine and bromine, and in all cases we may lay 
 down the law that the greater its affinity for chlorine, 
 bromine, or iodine, the greater the sensitiveness of the 
 
PHOTOGRAPHY. 5 
 
 silver-haloid. It should also be noted that however short 
 the exposure to light may be, the same changes occur in a 
 greater or less number of the molecules though such changes 
 may be invisible to the eye, owing to the preponderance of 
 the unaltered salts. 
 
 Now when Daguerre's plates were exposed in the camera 
 for a short time, no visible image was apparent, but, never- 
 theless, some minute quantity of the Ag 2 T 2 was converted 
 into Ag 2 T. It was found by Daguerre that such an invisible 
 image had the power of condensing mercury from mercury 
 vapour on the parts forming it, and that metallic lustre was 
 given to it. Thus the Daguerrean image of this white piece 
 of paper would have been represented by mercury and 
 sub-iodide of silver, while this black piece of paper would 
 have been represented by the silver iodide ; and when the 
 unaltered iodide is dissolved away, the latter would be re- 
 presented by the dark-coloured silver, and the former by the 
 lighter amalgam of silver and mercury. 
 
 I have here a glass plate silvered by Liebig's process, and 
 we will place it for a couple of minutes in this common 
 deal box, 1 at the bottom of which is a piece of cardboard 
 which has been exposed to iodine vapour during the night. 
 Iodine volatilizes at ordinary temperatures, so by leaving it 
 in the box the surface will be converted into silver iodide 
 by the combination between the metal and the halogen. I 
 now withdraw it, and examining it by candle light, 1 find it 
 of a delicate canary colour, with a slight tint of rose. It is 
 now in a sensitive condition to ordinary light. I will not 
 waste your time by exposing the plate in the camera, but 
 will place it behind a glass negative picture (what that 
 means we shall learn presently), and expose it for a few 
 seconds in the beam of the electric light. I believe it is 
 sufficiently exposed, so I will take it and hold it in the 
 vapour coming from the mercury (heated over a Bunsen 
 burner to about 150 F.) which is in this small capsule. The 
 image begins to spring out at once, and after a little longer 
 treatment it is fully developed. [The picture was handed 
 round.] 
 
 Now silver iodide, as I said, is sensitive to light, that is, 
 light changes it from the iodide to the sub-iodide so long as 
 
 1 The plate was supported on a couple of deal strips laid on the 
 card. 
 
6 LECTURES TO SCIENCE TEACHERS. 
 
 some body is present which can take up the liberated iodine. 
 What have we here, in this Daguerrean process, to do so ? 
 "We have the metallic silver, for recollect that the iodide is 
 only on the outer surface of the plate. Now I could demon- 
 strate to you the fact that this silver is necessary had I the 
 time. I might have repeated an old experiment, and, having 
 silvered a similar glass plate, and converted the whole of 
 the delicate metallic layer into silver iodide, have then shown 
 its insensitiveness to light, owing to there being nothing 
 to combine with the iodine, which it is anxious to liberate. 
 I will show you by-and-bye another experiment which will 
 illustrate the necessity of an absorbent. 
 
 Goddard, a countryman of ours, discovered that by treat- 
 ing the silver plate with bromine after the iodide had been 
 formed, the exposure in the camera necessary to form a 
 mercury-condensing image was shortened from minutes to 
 seconds. Perhaps this was the greatest of all improvements 
 in the daguerreotype process, as it rendered it thoroughly 
 practicable. 
 
 In 1834, whilst Daguerre was working in France on the 
 production of sun pictures, Fox Talbot, a gentleman whom 
 I am glad to say is still living, began experimenting 
 with silver chloride, pursuing the same line of thought as 
 Sir H. Davy, and in June 1839 (six months earlier than 
 the publication of Daguerre's process) he read a paper at the 
 Royal Society on photogenic drawing. This photogenic 
 drawing is really the same photographic printing process that 
 we employ now. Talbot impregnated writing-paper with 
 common salt or sodium chloride, and when dry treated it 
 with washes of silver nitrate, the result being to produce 
 silver chloride 1 in the paper with a little pure silver nitrate 
 ready to take up the chlorine which the darkening chloride 
 would liberate. Ferns, leaves, lace, &c., he copied by this 
 method ; more than rivalling the draughtsman in accuracy 
 and rapidity. 
 
 Let us suppose that one of the objects to be copied was 
 a piece of black lace. When the lace was laid on the paper 
 those parts beneath the cotton or thread would remain white 
 whilst the ground would be blackened. The paper on the 
 removal of the lace would represent the lace as white on a 
 black ground. This picture Talbot termed a negative picture, 
 i NaCl + AgN0 3 = AgCl + NaN0 3 . 
 
PHOTOGRAPHY. *, 
 
 or, as it is now called, a negative. Such a blackened paper 
 with the white image on it could be used to cover a second 
 piece of sensitive paper, which on exposure to light would 
 show the lace as black on a white ground ; and this he 
 termed a positive picture. Xote the advantage of this process 
 over the Daguerrean, which for each copy of an object 
 required a fresh exposure in the camera. 
 
 Hitherto we have dealt exclusively with the method of 
 
 Negative. 
 
 FIG. l. 
 
 producing silver iodide by the direct contact of the metal 
 with the halogen ; but the same results can be obtained 
 by chemical decomposition. Silver iodide may be precipi- 
 tated by mixing a solution of potassium iodide with silver 
 nitrate. 1 This was the method adopted by Fox Talbot in 
 the calotype process, patented in 1841. He added sufficient 
 potassium iodide to a solution of the silver nitrate to preci- 
 pitate silver iodide, and then an excess to redissolve it. Such 
 a solution he brushed over a piece of paper, which when dry 
 he washed, leaving on it primrose-coloured silver iodide. In 
 this state the silver compound was insensitive to light, as 
 there was nothing present (besides the paper) to take up 
 iodine. We have here a piece of paper prepared as in- 
 dicated, and it is now exposed to the strong glare of the 
 * KI + AgN0 3 = Agl + KNO,,. 
 
8 LECTURES TO SCIENCE TEACHERS. 
 
 electric light. To render it sensitive before exposure in the 
 camera Talbot brushed over it a solution of silver nitrate and 
 gallic acid, which I will get my assistant to do with our 
 prepared paper. The gallic acid plays a most important part 
 in the process, so much so indeed that I ought to mention 
 that its utility was discovered previously to the Talbot-type 
 process, by the Eev. J. B. Eeade, a gentleman who but a few 
 years since has passed away from amongst us. The method of 
 his discovery was systematic. When securing images in the 
 solar microscope he remembered that Wedgwood had produced 
 images on white leather and paper on which had been brushed 
 silver nitrate, and had found that his leather was more sensi- 
 tive than paper. It occurred to Mr. Eeade that the dressing 
 used for the leather might have some important property, 
 so he applied a solution of nutgalls to his paper, and found 
 the necessary exposure to light was greatly shortened. He 
 also discovered that this same nutgall solution had the power 
 of developing the image. This, however, was a chance dis- 
 covery. One day whilst engaged in producing photographically 
 an image of the Trientalis Europsea, he was compelled through 
 circumstances to put on one side the paper which had not 
 been sufficiently exposed to give a visible impression. Placed 
 in the dark, this paper was left till next day, when, on 
 glancing at it, he found a perfectly distinct image. The 
 gallic acid had played a part hitherto not dreamt of. 
 
 Eeverting to the calotype process, we find that Fox Talbot 
 employed the gallic acid with silver nitrate to render the 
 paper sensitive (the former being what is termed an ac- 
 celerator) ; and the discovery that an invisible image could 
 be rendered visible by the same solutions was also utilised, 
 for after exposure to the light the image was brought out by 
 them. 
 
 Our own piece of Talbot-type paper is ready, it having 
 been treated as indicated, and the excess of moisture blotted 
 off on blotting paper. After placing it behind the same 
 negative which we employed in illustrating the daguerreotype 
 process, we will expose it to the beam of the electric light. 
 A couple of seconds is a sufficient time to have produced an 
 invisible image, and we will at once proceed to render it 
 visible with a solution of gallic acid and nitrate of silver. 
 Tiiis we dab on with a tuft of white wool, and the picture 
 begins to appear. After a little patient manipulation the 
 
PHOTOGRAPHY. 9 
 
 whole of the details are brought out, and now we will place 
 it in a dish of water for a while. 
 
 This operation of causing the invisible to become visible, 
 how is it effected] We must set ourselves to solve the 
 problem by referring to a kindred action. If a rod of zinc 
 be placed in a strong solution of acetate of lead, by degrees 
 this latter becomes decomposed, and crystals of lead deposit on 
 the rod and completely cover it; but the action does not 
 cease when the covering is effected : the lead solution still 
 keeps depositing the metal, and a beautiful network of leaves 
 formed by metallic crystals is built up. The first particles 
 of lead deposited on the zinc attract other particles from the 
 solution till we have what is known as a lead-tree. 
 
 In the development of the image, as our last operation is 
 called, we have simply an example of the laws of crystal- 
 lisation, like crystals tending to adhere to like, and to be 
 attracted by them. Now in our exposed paper we had some 
 excessively minute portions of silver iodide (Ag 2 I 2 ) reduced 
 to the state of sub-iodide (Ag 2 I). Only one silver atom of 
 these more elementary molecules is saturated as it were, and 
 the other is free, and it is this free atom that is capable of 
 attracting metallic silver from a solution of silver nitrate, 
 when this latter is in an unstable state. In the present 
 instance the instability is caused by the gallic acid, for this 
 body tends to absorb oxygen, and as it absorbs oxygen it 
 liberates from the silver nitrate the metallic silver, and as 
 quickly as the separation is effected the free silver atoms 
 attract it. We thus get an image built up on the sub- 
 iodide ; for after one small particle of silver has been attracted, 
 it, in its turn, attracts others, as in the case of the lead 
 in the lead-tree. This then is the secret of development ; it 
 is the attraction exercised by the sub-iodide for freshly- 
 separated silver. 
 
 The developed image is therefore a metallic image ; but in 
 order to render it permanent, or perhaps I ought to say, more 
 clear, it was necessary to get rid of all the silver iodide. 
 To Sir J. Herschel belongs the discovery (in 1819) of the 
 solvent property of sodium hyposulphite on the silver chloride, 
 and it was by the application of this salt to the iodide that 
 the desired fixing of the image was effected. It is a matter 
 of surprise that this solvent was not employed at an earlier 
 date. 
 
10 LECTURES TO SCIENCE TEACHERS. 
 
 We will now take from the water our developed print and 
 fix it in a solution of sodium hyposulphite ; after washing 
 it will be permanent, or nearly so. 
 
 Ten years after the patenting of the calotype by Fox Talbot, 
 a new era arrived in photography. In 1851 was published 
 the collodion process a process which we use to the present 
 day, and one which there seems to be no chance of super- 
 seding, at all events for ordinary work. In the calotype 
 pictures the surface of the paper was found to be too rough 
 to render fine details, and at an early period of experimental 
 photography Sir J. Herschel had suggested the employment 
 of glass as a substitute, and in fact himself had produced 
 pictures on it, for in our Exhibition we find such a picture 
 taken as early as 1839. The method he adopted was to 
 obtain a fine precipitate of silver chloride in water, and at the 
 bottom of the containing vessel to place a glass plate. 
 After a lapse of some time the chloride was deposited with 
 sufficient solidity to render it practicable to remove the 
 glass from the vessel. After flowing over the crust of 
 silver chloride a little silver nitrate he allowed it to dry, and 
 exposed the plate in the camera. The picture I hand round 
 was produced in this manner. 
 
 At a later date Niepce de St. Victor went a step further 
 and employed a film of albumen for holding the sensitive salts 
 of silver in situ on glass an example of an early picture so pro- 
 duced is in the Exhibition ; but to Le Gray belongs the honour 
 of suggesting collodion as a vehicle to attain the same end. 
 Archer, with whom was associated Dr. Hugh Diamond, 
 however, practically introduced it. 
 
 Collodion is a solution of gun-cotton in ether and alcohol, 
 and when properly prepared should leave a transparent film 
 when the solvents evaporate. The iodides, bromides, and chlo- 
 rides of the alkalies, and also of many of the metals, are soluble 
 in alcohol, and can therefore be introduced into the collodion, 
 and be left in a film of this viscid body when poured over 
 a glass plate. A mere outline of the collodion process is as 
 follows : A plate is coated with collodion containing an iodide 
 and bromide, as I do this one [shows], and when the ether 
 has evaporated and the film is " set " or become gelatinous, 
 I place it in a dish containing a seven per cent, solution of 
 silver nitrate. On looking at the plate I see the silver 
 compound gradually forming, and after the lapse of about a 
 
PHOTOGRAPHY. 
 
 11 
 
 minute the film seems nearly opaque. I take our negative 
 picture, and, at each end of it lay a very thin strip of glass and 
 place the sensitive plate upon them. (The strips of glass 
 are used to prevent the surface of the collodion film being 
 abraded by contact with the other plate.) I expose to the 
 electric light for a second. On removing the plate I can see 
 no trace of an image, but I will endeavour to show you its 
 development. 
 
 Close to the lens (C) of the lantern (B) of the electric light 
 (A) I place a piece of red glass (E) in the clamp (D), for reasons 
 which subsequently will be apparent to you, and immediately 
 in /front of that a glass cell (F) containing a solution of 
 ferrous sulphate slightly acidified with acetic acid. I now 
 
12 LECTURES TO SCIENCE TEACHERS. 
 
 use a lens (G) to form an image of the cell on the screen. 
 This being arranged, I dip the glass plate into the solution. 
 You see at first not a sign of any image, but only a semi- 
 opaque film. But now we see a darkening in parts, and a 
 picture is gradually appearing. It gains intensity, and now 
 it is perfect in all its details. We will withdraw it and 
 wash it, and then treat it with a little potassium cyanide, 
 which is also a solvent of many of the compounds of silver. 
 After washing again we will place it in the lantern and throw 
 the finished picture on the screen. 
 
 Here then we have a picture produced by the collodion 
 process. You will have noticed that this time the image 
 was brought out by a solution of ferrous sulphate and not of 
 gallic acid. Ferrous sulphate is a greedy absorber of oxygen, 
 and therefore is effective in causing a reduction of silver from 
 the nitrate, with which the plate was impregnated before 
 placing in the cell. 
 
 We are now in a position, I think, to make an experiment 
 which I promised at an earlier part of the lecture, viz., to 
 prove to you that silver iodide is unaltered by light unless it 
 has some iodine-absorber present with it. In my hand I hold 
 a glass plate to which is adhering a collodion film containing 
 pure silver iodide and having no excess of silver nitrate. 
 After placing it in the direct rays of the beam from the electric 
 light, I apply a solution of pyrogallic acid and silver nitrate 
 to it, and there is no change apparent. Taking a similarly 
 prepared plate, I apply a small square piece of silver leaf to 
 it, brushing it well on to the film. With this camel's hair 
 brush on another portion I brush a solution of tannin in 
 alcohol, and after warming the glass through its back to 
 cause desiccation, Iwill expose it for half a minute to the light, 
 behind a negative. We will develop it in the same way as 
 our last picture, using a solution of pyrogallic acid, however, 
 instead of ferrous sulphate. Notice the result only those 
 portions of the negative appear on our plate which have been 
 coated with the silver or have received the wash of tannin. 
 After fixing our picture I throw it on the screen, and we see 
 our results more perfectly. The transparent parts through 
 which the light passes show where the actinic rays did not 
 affect the silver iodide. 
 
 You will notice that the part A is somewhat fainter than 
 B. The former you will recollect was the part on which was 
 
PHOTOGRAPHY. 13 
 
 placed the silver leaf, whilst to B was applied the tannin. 
 The reason of the difference is obvious. It is only those 
 particles of the silver iodide in actual contact with the 
 iodine-absorber which can be affected by the light. In the 
 
 FIG. 4. 
 
 one case the silver leaf was only in contact with the surface 
 particles, in the other the tannin permeated the film. 
 Having thus prepared the ground, we are in a position to 
 understand the preparation of dry plates, as they are called, 
 that is, the preparation of sensitive films in collodion which 
 can be exposed in the dry state. 
 
 If a surface of silver iodide were allowed to dry with the 
 free silver nitrate solution from the bath upon it, the latter 
 would crystallize and spoil the film. If we wash away 
 the free silver nitrate it would be insensitive, but if 
 
14 LECTURES TO SCIENCE TEACHERS. 
 
 in its place we applied some iodine absorbent (after washing] 
 which is less crystallizable, such as tannin (as we did in our 
 last experiment), salicin, tea, coffee, &c., it will again be sensi- 
 tive even when dry. The application of such bodies is the 
 foundation of any dry plate process in which the iodide plays 
 a part. At the same time these absorbents are given to the 
 iodide, there are usually added compounds which soak into 
 the film and keep the pores of the collodion open. After 
 washing the exposed film these of course dissolve away, and 
 allow free access to the developer. The same procedure holds 
 good when both bromides and iodides are present in the sen- 
 sitive film, and we may develop with iron or pyrogallic acid 
 and silver, as in the wet collodion process, building up the 
 image on the irritated haloids. 
 
 When silver bromide is present alone, or in conjunction 
 with the iodide, however, another method of development 
 may be resorted to, viz., one that is primarily independent of 
 the deposition of metallic silver. 
 
 Pyrogallic acid has an affinity for bromine as well as for 
 oxygen, and the affinity for both is multiplied many fold 
 by the addition of an alkali to it. Pyrogallic acid and an 
 alkali such as ammonia (say) form the pyrogallate of the 
 alkali. When such a solution is poured over the exposed 
 bromide film it meets with molecules of the sub-bromide 
 (Ag 2 Br) which appear to be in an unstable state, and ready 
 to part with any of their atoms. The pyrogallate speedily 
 separates the remaining atoms of bromine from them, and 
 leaves the metallic silver behind. When ammonia is 
 the alkali employed with the pyrogallic acid, we have a 
 further action. Over this glass plate, on which is a layer 
 of silver bromide in collodion, I allow a few drops of 
 strong ammonia to trickle. Notice the transparent track they 
 leave. From this we learn that the bromide is soluble in 
 ammonia. Hence, whilst the pyrogallate is still acting in 
 the manner indicated, if there be sufficient ammonia present 
 it dissolves a portion of the silver bromide ; this the pyro- 
 gallate decomposes, and causes the metal to deposit as in 
 the ordinary development. By this means a greater density 
 is given to the image than would otherwise exist. There 
 is a further density given by a somewhat curious action 
 (apparently catalectic) the cause of which I have not time to 
 explain. 
 
PHOTOGRAPHY. 15 
 
 I should state that sometimes the avidity of the pyro- 
 gallic acid in the presence of ammonia for bromine is so 
 great that it is found necessary to give it a soluble bromide 
 wherewith to satisfy it. Thus potassium bromide is usually 
 added to the alkaline developer. We will expose a plate 
 prepared with silver bromide behind a negative and develop 
 it by this alkaline method, throwing the resulting picture 
 on the screen; There is no noticeable difference between 
 this image and that developed with the ferrous sulphate in the 
 wet plate. 
 
 Now it is not necessary to prepare a sensitive bromide 
 film by immersing the collodionized plate in a solution of 
 silver nitrate ; by an artifice we can have the sensitive salts 
 held in suspension in the viscid collodion. 
 
 I will very briefly cany you through the operations 
 necessary to produce this emulsion, as it is technically called, 
 of silver bromide. In collodion are dissolved soluble bromides. 
 Silver nitrate dissolved in weak alcohol is added in sufficient 
 quantity to convert it into silver bromide. If rightly 
 carried out, this solid bromide remains in suspension in a very 
 finely divided state in the collodion. The collodion is next 
 poured out into a dish and allowed to become gelatinous by 
 the evaporation of the ether, after which it is well washed 
 (to eliminate all the soluble salts present), and dried. 
 The pellicle is next redissolved in a mixture of ether and 
 alcohol, and we have as a result such a viscous fluid as in 
 this bottle. To prepare a film for exposure in the camera, 
 all that is necessary is to pour it over the surface of a glass or 
 other plate as if it were ordinary collodion. It is equally 
 well acted on when dried as when still moist with the 
 solvents. It is usual to add some bromine absorber to the 
 collodion, but this is not absolutely necessary, though it is 
 generally considered that sensitiveness is increased by so 
 doing. 
 
16 LECTURES TO SCIENCE TEACHERS. 
 
 PHOTOGRAPHY. 
 
 LECTURE II. 
 
 WE left off with the development of the dry plate by the 
 method known as the alkaline process, and this morning 
 we must break into new ground. But yesterday after the 
 lecture a gentleman came to me, and said, "You have not 
 spoken about the latent image." Now what is called the 
 latent image is a thing I hesitate to recognise. I will admit 
 there is an invisible image, but . not what should strictly be 
 called a latent image. If we take 10,000 parts of oxide 
 of zinc, say, and mix thoroughly with it one part of lamp- 
 black, you will perceive no difference in tint between the 
 pure oxide of zinc and that contaminated with the lamp- 
 black ; but the lampblack is there nevertheless. So in the 
 same way when we expose for a short time a sensitive plate 
 to the action of light, a change has taken place in certain 
 molecules exactly in the way that would give us a visible 
 image by a longer exposure, the only difference between 
 the one and the other being in the quantity of molecules 
 affected. I think I need scarcely repeat the experiments 
 which showed the development of the invisible image. 
 Suffice it to say that the visible as well as the invisible 
 image, can attract silver from an unstable solution of silver 
 nitrate. 
 
 But to-day we have to consider as to the light which is 
 most favourable for photographers, i.e. to what rays of light 
 these salts of silver are most sensitive. If you cause a slice of 
 white light to pass through a prism, it is separated into 
 its component rays, and we have what we all know as the- 
 spectrum. The spectrum of the light from the incandescent 
 carbons is now thrown on the screen, and if I were to place 
 a large piece of paper impregnated with a sensitive silver 
 salt such as the iodide in that spectrum and I have taken the 
 iodide with a set purpose we should notice, on the appli- 
 cation of a developing agent, that no change in its molecules 
 had been effected by the red and yellow rays, but that green, 
 blue, violet, and the invisible rays beyond the violet 
 
PHOTOGRAPHY. 17 
 
 had "been active. The length of the ultra-violet part of 
 the spectrum is equal in length to the whole of the 
 visible spectrum. Mr. Lockyer and Professor Stokes, I 
 think, have told you about these ultra-violet rays, and I am 
 not going to repeat what they have said. Instead of using the 
 electric light we may pass a very thin slice of sunlight 
 through a prism or a couple of prisms, and when we do so, 
 we find a spectrum traversed by black lines, which you have 
 already heard are due to the absorption of metallic and 
 other vapours. Those lines, as you are aware, occupy certain 
 fixed positions in the spectrum, and supposing that we get 
 photographic impressions bounded by any particular line, we 
 should know what part of the spectrum was effective. 
 
 Before proceeding further I may show you that with the 
 ordinary silver salts employed the red light is inoperative to 
 produce a picture, whilst the blue light is perfectly capable 
 of so doing. I have here a dry plate prepared with 
 bromo-iodide of silver (I told you yesterday that we mixed 
 bromides with iodides in the collodion, and produced silver 
 bromo-iodide in the film by means of a solution of silver 
 nitrate), and I will expose, for a minute, one half of it behind 
 a negative to the red rays, and the other half for ten seconds 
 to the violet rays. You will see that in the first half of the 
 picture we shall get no results, and in the other half we 
 shall get an image. After it is developed I will throw it on 
 the screen. Here it is after the developer has been applied, 
 and you see the red light is incapable of impressing an image. 
 Before I came yesterday I photographed the spectrum on 
 different silver compounds, to show what rays are capable of 
 producing an image, and when you compare this photograph, 
 taken on the iodide salt, with the carefully coloured drawing 
 hanging on the wall, you will find what I told you was 
 correct about the silver iodide. [The photograph shown 
 on the screen.] The transparent line, which you see at 
 the extreme end of the image, agrees with the line E of the 
 solar spectrum, and when we turn the light on the diagram 
 of the spectrum you will see what position this line occupies 
 in the green. The spectrum goes along through the blue, 
 through the violet, and here we get two well-known lines, 
 called the H lines, also to be seen in the diagram, which 
 are very near the extreme end of the violet portion. You 
 can now see that the photographic action goes far beyond them, 
 
18 LECTURES TO SCIENCE TE AGREES. 
 
 right into those portions which are usually invisible. To 
 obtain this photograph and those which I shall show you 
 directly, the spectrum was thrown on the sensitive com- 
 pound after passing through ordinary glass prisms and glass 
 lenses. If I had employed quartz lenses and prisms, 
 or Iceland spar lenses and prisms, I should have been 
 able to obtain an impression much further in the ultra- 
 violet, because glass cuts off these rays to a great extent ; 
 but as photographers use glass objectives, I thought it best 
 to show you the spectra as produced through this medium. 
 
 I now wish you to compare the spectrum when photo- 
 graphed on other silver compounds, with that already shown. 
 You will notice that the iodide shows the greatest impressi- 
 bility to those rays which correspond to G, whose wave- 
 length answers to about 4,300 metres, 1 at which point 
 
 it seems to tumble down a precipice in the direction of 
 the violet. The lowest ray which effects it is the E, which 
 corresponds to about 5,200. The maximum effect produced on 
 the bromide seems to take place at about the same wave-length 
 and to diminish more gradually and regularly in both directions. 
 On the chloride the intensity seems to be nearly of the 
 same character as that of the iodide, though the fall in 
 effectiveness is not so marked towards the H line. We may 
 say, then, the gradients of the sensitiveness of the bromide 
 to the spectrum are far less steep than those of the iodide 
 and chloride. The question comes, then, is it not possible 
 to have some compound which shall give a gentler gra- 
 dient than the bromide, and thus enable us to photograph 
 further in both directions, or, if that be not possible, cannot 
 we change the point of maximum effect to a point nearer 
 towards the red end, by employing a different silver com- 
 pound, and yet preserve the same gradient, as the bromide ? 
 In either case we should be able to photograph further down 
 towards the A line. 
 
 Each ray of light, as you doubtless are aware, is caused by a 
 different vibratory motion of the all- space-pervading ether. 
 The waves producing the red rays are longer than those pro- 
 ducing the orange, the orange than the green, and so on. Now 
 
 1 In future we shall refer all these, wave-lengths to the same scale 
 
 of metres ' 
 
PHOTOGRAPHY. 
 
 FIQ. 5. 
 
 C2 
 
20 LECTURES TO SCIENCE TEACHERS. 
 
 when any one of these rays impinges upon a molecule of 
 matter, it meets with an opposition to its motion ; it may be 
 that the swing of the wave may be in accord, or nearly so, 
 with the natural swing of the molecule. In such a case we 
 might expect no change to occur in a compound molecule 
 though motion be imparted to it. But it may happen that 
 the beats of the two are of such a nature that a violent 
 internal battering and sifting, as it were, of the molecule will 
 take place. It may be so driven and shaken about, that in 
 order to arrive at some sort of harmonious motion with the 
 wave, it may throw off one of the atoms of which the 
 molecule is composed. It seems probable that to pro- 
 duce photographic sensitiveness in a compound there are two 
 requirements, first, that the molecule should be set in motion 
 by the wave, second, that the motion must be of such a 
 nature that a sifting of its component atoms takes place. 
 The silver iodide 1 molecule is apparently in the greatest 
 internal discord with the blue ray waves, and when they 
 fall upon it, it throws off an iodine atom. If then we can 
 obtain a molecule which can be caused to vibrate by the im- 
 pact of the red ray waves, and yet be shaken by them, we 
 might expect that such a one would throw off a something 
 to render itself more in tune with that wave. 
 
 I give beneath a table of comparative weights of the 
 different molecules of silver chloride, bromide, and iodide. 
 
 Ag 2 Cl 2 Ag 2 Br 2 A 2 I 2 
 
 287 376 470 
 
 By loading any of these silver compounds we ought to be 
 able to produce a corresponding change in the limit of internal 
 discord. Thus by loading a silver chloride molecule with a 
 dead weight equal to 89, or the difference between 376 and 
 287, we ought to cause its limit to be the same as the bromide. 
 Again, we have every reason to believe that by slightly 
 loading the molecule of the bromide we might make its limit 
 of internal discord to be at the A line, or below it ; or again, 
 we may expect that by getting a compound of silver which is 
 heavier than 470, the molecular weight of silver iodide, we 
 might obtain similar results. In all cases we must suppose that 
 there is some portion of the molecule that can readily be 
 
 1 The absorption spectrum of silver iodide as well as of bromide 
 shows the greatest absorption to take place in the blue. 
 
PHOTOGRAPHY. 
 
 21 
 
 shaken off by it when the swing of the wave is sufficiently- 
 inharmonious. It should be noted, however, that in these 
 loaded molecules the increased weight may be a dead weight, 
 as it were ; that is, the part shaken off may still be chlorine, 
 iodine, or bromine, if the compound contain any or all of 
 these halogens. I now throw on the screen a photograph of 
 the spectrum taken on loaded silver bromide. The bright 
 line in the centre is the extreme limit of visibility, the A 
 line ; on the left beyond the A line there are indications of 
 bands of lines, and when you come to examine microscopically 
 a good photograph, you can see that it is particularly rich in 
 lines, a great many of which are due to the absorption of 
 the atmosphere. 
 
 FIG. 6. 
 
 To show you that it is not only the solar spectrum that 
 can be photographed by this means, I have here the first 
 photograph taken by myself of the red end of a spectrum, 
 of a metal, namely, of calcium. This metal is particularly 
 rich in red rays, together with some rays beyond the extreme 
 limit of visibility. I may remark in passing that by the 
 process here adopted the point of maximum photographic in- 
 tensity in the spectrum is lowered towards the red. Those 
 two black lines on the left are the last which can be seen, 
 and the lines towards the right lie beyond the A line. Their 
 waves are of such a length that they cannot impress them- 
 selves on the nerves of the eye. 
 
 I should like to expose a plate coated with this sensitive 
 salt, to show you that there is no illusion about photo- 
 graphing with red rays. I have here such a one which I will 
 expose to the red light only ; it will require perhaps a 
 minute's exposure, and I must have it developed as far as 
 possible in the dark, otherwise the image would be veiled. 
 [The developed image was then thrown on the screen.] 
 
 Dr. Vogel, of Berlin, has experimented with bromide 
 plates prepared in the silver bath, which, after washing, he 
 nooded with various diluted dyes. He has found and often 
 
22 LECTURES TO SCIENCE TEACHEES. 
 
 correctly, that the film becomes sensitive to the action of those 
 rays which the dyes will absorb. Blue, however, according to 
 his theory, should absorb all the red rays, and thus be chiefly 
 sensitive to the red and yellow. Experiment does not always 
 bear him out in this. I am inclined to think that by the 
 use of these dyes he forms a real compound with silver (for 
 however well you may wash your plates you cannot eliminate 
 all free nitrate of silver), and thus loads his molecules. To 
 explain Vogel's reasoning I will throw the spectrum on the 
 screen, and place a glass cell containing cosine dissolved in 
 alcohol before it ; you see there is an absorption in the green, 
 whilst the fluid allows the rays immediately above and below 
 it to pass. 
 
 You must bear in mind that where absorption occurs, there 
 work of some description must be performed, for the diminution 
 of the amplitude of a wave denotes that energy has been 
 expended. 
 
 If we faintly stain a film of silver bromine with eosine, so 
 that the dye absorbs the green rays through its whole thickness, 
 and carries, as it were, the light to every individual molecule 
 in its path, the rays not so rapidly absorbed should not 
 cause such a rapid chemical change. By increasing the 
 strength of the dye on the washed plate it is possible to 
 prevent any change in the portion exposed to the green 
 rays, as they are absorbed before they get to the bromide, 
 but if at the same time we increase the silver in the film to 
 such an extent as to allow the whole of the dye to form 
 a compound with it, the effect is still more marked. By 
 this procedure I have been enabled to photograph below 
 the "a" line, with the weighted molecule of bromide. 
 I throw on the^ screen a spectrum taken on a plate 
 slightly stained with eosine (see Fig. 5), and I will ask you to 
 compare it with the absorption spectrum of the dye and also 
 with the spectra impressed on the other silver compounds. 
 As I said before, I believe that the results obtained with the 
 dyes by Vogel are due to their combination with the residue 
 of silver nitrate left in the film, and what seems to tend to 
 confirm this view is that if the film contain excess of soluble 
 bromide, no results are to be obtained. 
 
 Dr. Vogel's researches do not absolutely point to obtain- 
 ing a method of prolonging the photographic spectrum, 
 yet they tend towards it. You must not accept my explana- 
 
PHOTOGRAPHY. 23 
 
 tion of them as correct (though I believe it to be so), as the 
 theory is still in the course of examination by Dr. Vogel, 
 Captain Waterhouse, myself, and others. 
 
 We must again travel forward a step, and now we shall 
 find it more easy to understand the production of photographs 
 of a certain kind which we find in the Exhibition. If we 
 take pure silver chloride and expose it to the action of the 
 spectrum sufficiently long to give a visible impression, we 
 find that it extends from the green to far beyond the. extreme 
 violet, but that no action takes place in the red. I exhibit 
 here a photograph of the spectrum taken on paper impregnated 
 with silver chloride ; the exposure necessary to produce this 
 print was three-quarters of an hour. Another similar piece of 
 paper was taken and slightly darkened in diffused light, and 
 placed in the spectrum. The impression of the blue rays 
 continued, and at the same time there was a browning action 
 taking place where the yellow rays were thrown, whilst at the 
 extremity of the red a decided pink was apparent. Two 
 such spectra I have here, one fixed and the other not fixed ; 
 you will perceive that in the former the red tint has been lost, 
 though there is an evidence of decided darkening beyond 
 that due to tint on the paper. In the latter the colours are 
 still extant. 
 
 Now conjointly with these results I wish to bring to your 
 notice another experiment that any of you possessing a prism, 
 and a camera, and a looking-glass can repeat, if you have 
 even the smallest photographic knowledge. Take a glass 
 plate and prepare it as if you were going to take a picture, 
 expose it to the daylight and develop it, then "intensify" 
 till you have a perfectly opaque film of silver, and dissolve 
 away the unaltered silver salts, and wash well. Next 
 take some copper chloride and flood the film with it. It 
 will gradually seem to turn to a dirty white by reflected 
 light, and by transmitted light it will be a reddish brown, 
 for the copper will part with half its chlorine to the silver, 
 which there is every reason to think is converted into a 
 mixture of chloride and sub-chloride. Now expose such a 
 plate to the action of the spectrum for ten minutes. You 
 will find that where the violet, blue, and green rays come, 
 you have a darkening of the surface, whilst where the 
 yellow and red rays fall you have a bleaching action. 
 
 Again, take another plate, similarly prepared, but give 
 
24 LECTURES TO SCIENCE TE AGREES. 
 
 it a final wash with silver nitrate, and allow it to darken 
 in white light, and then expose it to the spectrum, or 
 "beneath red, green, yellow, and blue glasses. In this 
 case we shall find the silver reddens in the red light, becomes 
 greenish in the green, and blue in the blue. I have here a 
 plate that has been so treated and exposed to the different 
 coloured lights, and you can note the colours though they 
 are somewhat spoilt by subsequent experiments. Mark the 
 difference between the two plates : one had a chloride and 
 sub-chloride of silver alone, the other had chloride, sub- 
 chloride, and a chlorine absorbent present. 
 
 We may class the result obtained on the silver chloride on 
 paper arid that on the last coloured plate as identical, both 
 being exposed under the same conditions. Leaving out, for 
 the present, the theory of the production of the colour, let us 
 examine the fact that a change has been effected by the rays 
 of low refrangibility. We have silver chloride and sub-, 
 chloride together in close contact on the surface, and I think 
 we may take it that we have the case of a loaded molecule, 
 whose swing is in discord with the longer waves ; that such a 
 discord causes an atom of chlorine to be thrown off (as before 
 explained), and that the atom so separated is from the portion 
 of the compound molecule which by itself would be the sub- 
 chloride. As a result of the impact of the light we should 
 have remaining metallic silver (from the sub-chloride) and 
 unaltered silver chloride. The amount of sub-chloride formed 
 by the preliminary exposure would be small, hence the total 
 amount of reduced silver would be little in comparison with 
 that which would be due to the reduction of the chloride to 
 the sub-chloride by the more refrangible portion of the 
 spectrum. The well-known reversal of the lines of the red 
 end when photographed on iodide or bromide of silver, to 
 which a slight preliminary exposure has been given, can 
 be accounted for in the same way, on the supposition that 
 the silver reduced from the sub-iodide or bromide is not so 
 actively attractive as the sub-iodide or sub-bromide itself. 
 The fact that such reversed photographs are always more or 
 less veiled is rather confirmatory of this view. 
 
 With the plate treated with copper chloride alone, and no sub- 
 sequent addition of silver nitrate and preliminary exposure, 
 the same line of argument still holds good. Part of the sub- 
 chloride is reduced to silver, and chlorine is evolved, the latter 
 
PHOTOGRAPHY. 25 
 
 being absorbed by another portion of the sub-chloride, with 
 which it combines to form the white chloride. The minute 
 atoms of reduced silver are shrouded by its whiteness, and 
 we have the consequent appearance of the bleaching of the 
 brownish-coloured film. 
 
 The cause of the colours in the paper-print and in the 
 plate requires explanation. The fact that when the unaltered 
 compounds on either of them are dissolved away, the colour 
 vanishes, leaving only that due to silver itself throws a light 
 on the subject. 
 
 In a soap bubble the beautiful colours which overspread 
 its surface are caused by the interference of the light re- 
 flected from the outer and inner surfaces which are micro- 
 scopically near each other, and it may be that the colours pro- 
 duced by the spectrum are the results of the interference of 
 the light reflected from the surface of the reduced particles 
 which are held apart by intervening silver chloride. 1 After 
 dissolving out the latter, the particles are brought in contact 
 and the colour disappears. 
 
 In the Exhibition we have photographs in colour, re- 
 presenting the solar spectrum, by Becquerel, and one of 
 my objects in leading up so far as I have was to try and 
 give you an explanation of the method of their production. 
 On a bright silver surface silver sub-chloride was formed by 
 voltaic or other means, and a spectrum was caused to fall on 
 a plate so prepared imprinting itself in all its colours. If 
 exposed to the light these spectra fade away and leave nothing 
 behind but a bluish brown plate. Hence it is that they are 
 preserved in closed light-tight cases, and can only be rarely 
 exhibited. 
 
 The cause of an action taking place by the impact of the 
 red and yellow rays has already been pointed out ; and the 
 vividness of the colours can also be readily accounted for 
 by the same explanation as given to account for those on 
 our paper and collodion film, when it is remembered that 
 there is the reflecting surface of the silver plate itself to aid 
 the interference. 
 
 Hitherto we have only spoken about silver compounds 
 being sensitive to light, but nearly all matter is sensitive 
 in one respect or another. Most probably the first action 
 
 1 Silver chloride is really a white transparent substance, as may be 
 proved by fuzing it in a crucible. 
 
26 LECTURES TO SCIENCE TEACHERS. 
 
 with, which man was acquainted as photographic action was 
 that of the tanning due to the sun, and in very primitive 
 days no doubt it was more marked than it would be at 
 the present day. The next change most probably would be 
 noticed by the fair sex, who used coloured materials for their 
 dresses, and ladies soon found out that silks, calicoes, or 
 ribbons of certain colours materially changed under the action 
 of light. Here I have three pieces of different coloured 
 materials^ on which you are able to produce an image by 
 the fading of the dyes. These have been exposed under a 
 negative to the action of light, not heat. The usual explana- 
 tion about the fading of these colours is that there is something 
 given off like the scent from a rose ; but when you see a map 
 absolutely printed by light on them, you can have no doubt 
 as to the action which has produced it. If it were that a sort 
 of essence is given off, a negative placed over these 
 colours whilst in the light would produce no defined result 
 whatever, they would fade equally under it ; we therefore 
 cannot help concluding that some chemical change has taken 
 place. Again, you will find most unlikely substances, such 
 as glass, change under the influence of light. My friend Mr. 
 Dallmeyer has some beautiful specimens of glass, which 
 have been altered in this way ; flint glass being changed 
 to a yellow colour and crown glass to a purple tint. Again, 
 we know that there are elements which are affected by light, 
 and amongst them I may mention selenium, a body whose 
 resistance to the passage of a current of electricity it has been 
 proved is diminished by the impact of light. Thus in 
 darkness a piece of selenium 1 '5 in. X '5 in. x '05 in. offered 
 333000 units of resistance to the passage of the current. 
 Whilst in the diffused light it offered nearly 270000 units. 
 It was also found that the resistance was decreased most 
 in the least refrangible portion of the visible spectrum. In 
 the blue it was only 279000 units, in the yellow 277000 
 units, and in the red 255000 units of electrical resistance. 
 I have brought this forward to show to you that a simple 
 elementary body may be acted upon by light. 
 
 We also find that the colouring matter of flowers and of 
 leaves is affected by light. Mrs. Sornerville and Sir John 
 Herschel made a long series of experiments with it. If you 
 take the leaves, say, of common cabbage, and place them in 
 alcohol, a certain coloured resin is extracted, which is known 
 
PHOTOGRAPHY. 27 
 
 by the name of chlorophyll. Its solution is red by trans- 
 mitted, and green by reflected light. Here we have a piece 
 of paper which has been brushed over with this alcoholic 
 solution : the colour is a sort of primrose green. By exposure 
 to light it has become bleached. This bleaching is principally 
 due to the yellow light and not to the blue light, which 
 acts on the salts of silver. Again, if you take the leaves of 
 stocks, common wallflowers, violets, or roses, and treat them 
 with alcohol, you can extract the colouring matter, and if, 
 having brushed it over, you expose it beneath a negative, you 
 will get prints of various colours. If you treat the rose extract 
 with a small quantity of acid and brush it over a sheet of 
 paper and expose it to the light, you will find the natural 
 pink colour intensified and the subsequent change will be 
 increased. Again, take the common violet, treat its extract 
 with ammonia, and it gives you a green solution, but the 
 green colouring matter is bleached by the action of light, 
 and experiment proves that the parts of the spectrum to 
 which the colouring matter is sensitive are not the same 
 as those to which the silver salts are sensitive. There is 
 a wide range of experimental work yet to be undertaken 
 with respect to this colouring matter of flowers. 
 
 I should here like to call your attention to the fact that 
 some gaseous bodies as well as solids are affected by light. If, 
 for example, we take hydrogen and chlorine in proper propor- 
 tions in a glass bulb, and keep them in the dark, no com- 
 bination takes place ; but if we take such a bulb into sunlight, 
 they combine almost instantaneously, the light causing the 
 atoms to swing in such a way that they mutually attract 
 each other, and form hydrochloric acid. As Dr. Tyndall has 
 shown, it is not heat-waves that cause the atoms to com- 
 bine, but light- waves ; he enclosed the gases in a collodion 
 balloon, and then caused them to combine by a concentrated 
 light, and the film was found unburnt. In diffused light 
 the combination takes place slowly and without explosion. 
 For similar reasons, if chlorine be passed into water in the 
 daylight, hydrogen is abstracted from the water and hydro- 
 chloric acid is formed, the oxygen forming another compound 
 with the chlorine. 
 
 The bodies to which I next shall call your attention as 
 sensitive are metallic compounds. Sir John Herschel was 
 the first to investigate the action of light on iron com- 
 
28 LECTURES TO SCIENCE TEACHERS. 
 
 pounds, and to him are due a variety of most interesting 
 processes, examples of one or two of which. I shall endeavour 
 to show yon. 
 
 Whilst mentioning the above great philosopher, I should 
 like to point out to you the instrument with which he 
 operated when experimenting on the effect that light pro- 
 duced on different organic and metallic compounds. 
 
 FIG. 7. 
 
 A is a glass prism which could be rotated in its frame 
 round an axis D, so that the sun's rays would be dispersed 
 in any given direction. B is a lens denning the spectrum 
 (which, it must be recollected, was a mixed one, and not 
 pure). C the screen on which the spectrum was raised, and 
 on which the compound to be tried was placed. At E was 
 marked a line on which one particular part of the spectrum 
 was invariably caused to fall. When used the frame-work 
 was covered with a black velvet cloth. The absolute results of 
 the experiments on the different compounds are shown in an 
 old book containing a list of Fellows of the Royal Society, 
 over whose distinguished names they have been pasted. 
 This book is perhaps one of the most interesting exhibits 
 in the Loan Collection. 
 
 Reverting to the iron salts I may point out that those of 
 them which are in the ferric state are the most readily acted 
 upon by light ; the ferrous salts, as a rule, not being sensi- 
 tive. A variety of ferric salts may be formed, such as ferric 
 chloride, or ferric oxalate ; or you may have a compound of 
 ferric citrate with citrate of ammonia, and so on. Of all 
 the compounds of iron, Sir John Herschel found that this 
 latter ferric salt, when employed with the ammonium citrate, 
 
PHOTOGRAPHY. 29 
 
 was the most easily operated upon. If we take a piece of 
 paper and brush over it a mixed solution of these two salts, 
 and when dry expose it to light beneath a negative for a short 
 time (about two minutes), we shall obtain a blue image after 
 treating it with a solution of potassium ferri-cyanide>. This 
 blue colour can only result from the contact of the potassium 
 ferri-cyanide with some ferrous compound. Here then we 
 have a demonstration of the change effected by light ; the 
 ferric compound is reduced to a ferrous state. 
 
 I will endeavour to produce such a print before you, but I 
 must tell you that these iron salts are most objectionable for 
 lecture experiments. If you expose a piece of paper which 
 has been coated with this ferric compound, causing an image 
 to be formed of a ferrous salt, and put it away in the dark, 
 it rapidly loses the impression altogether. The ferrous 
 becomes reconverted into the feme salt. This is exceed- 
 ingly tantalizing. I had some sheets of prepared paper 
 exposed only this morning, and on developing them by the 
 method already indicated, you see the image is very weak, 
 due to this reactionary cause ; had it remained undeveloped 
 a few hours longer we should have had no image at all. 
 
 An iron print can also be developed by means of silver 
 nitrate. A ferrous salt of iron will reduce silver nitrate 
 to its metallic state, as already shown in yesterday's lecture ; 
 and if I bring a solution of the latter on to the exposed 
 print, you will see that the silver deposits on those parts 
 affected by light. 
 
 The next salts to which I must call your attention are the 
 uranium salts, of which there are specimens on the card 
 which you see before you. Uranium nitrate is sensitive to 
 light in he presence of organic matter, being thus reduced 
 to the state of an oxide. This oxide precipitates silver and 
 other metals from their solutions, and with potassium ferri- 
 cyanide forms a nearly insoluble brown compound. 
 
 We will now develop a picture with the ferri-cyanide, and 
 you will note its appearance. I have also another photo- 
 graph printed with uranium, which is now being placed in 
 a solution of, silver nitrate, to which a little gallic acid 
 has been added. The silver is gradually being reduced by 
 uranium oxide, and a metallic image is being built up. After 
 passing the paper through sodium hyposulphite and wasliing, 
 the picture is permanent. 
 
30 LECTURES TO SCIENCE TEACHERS. 
 
 The other most interesting compounds to which I would 
 call your attention are those of vanadium. Professor Eoscoe 
 found, during some recent researches, that certain vanadium 
 salts were sensitive to light. Here is the first vanadium 
 print ever produced. It was developed by silver nitrate 
 in a manner similar to that employed with the uranium print. 
 
 I am obliged to pass over some other metallic compounds, 
 but 1 must mention the potassium dichromate, or rather 
 the chromium salts. These salts are the great handmaidens 
 of photographic printing processes at the present day. When 
 you brush a solution of potassium dichromate over paper and 
 expose it to the light, you will find the paper becomes 
 darkened where the light has acted, and an oxide of chromium 
 has been formed, the organic matter in the paper having 
 reduced the potassium dichromate to that state. I have here 
 such a piece of paper which has been exposed under a 
 negative, and washed afterwards ; and you see the green 
 coloration of the chromium oxide. 
 
 Not only is the potassium dichromate reduced to the state 
 
 FIG. 8. 
 
 of oxide, but it also oxidises the organic matter with which 
 it is in contact. If you take gelatine or any similar colloid 
 body and add fco it a solution of potassium dichromate, dry 
 
PHOTOGRAPHY. 31 
 
 it, and expose it to the light, you will find the gelatine has 
 undergone a distinct change. First of all, the gelatine has 
 become insoluble in hot or cold water, and in the second 
 place it has become incapable of absorbing water. I have 
 here a sheet of paper (A) coated with gelatine, in which was 
 dissolved potassium dichromate. It has been exposed to 
 light under a negative of a map, and, as I have just told 
 you, it will not absorb water where the lines are printed. 
 
 I immerse it in \vater, place it on this glass plate (B), blot off 
 the excess of moisture, and then roll it with this roller (C), 
 which has been charged with greasy ink. The ink is already 
 beginning to take on the lines ; I run the roller briskly over 
 it once or twice to give more ink to them and to remove any 
 adhering superficially, and we have now got a finished map 
 whose lines are formed of the greasy ink ; this process of 
 reproducing plans I have called the papyrotype process. 
 
 In a simple way I want to show you also that gelatine 
 becomes insoluble in hot water by the action of light when it 
 is in contact with potassium dichromate. I have here a piece 
 of gelatinous paper, which has been exposed under the same 
 negative as before, its surface whilst dry has been covered with 
 a thin layer of greasy ink. Now if gelatine becomes insoluble 
 where the light has acted, when I float this paper on hot water, 
 those parts which have been acted upon by light ought 
 to remain on the paper, and those parts which have not 
 been acted on ought to dissolve, carrying the ink with 
 them. I place the uncoated side of the paper on the boilin-g 
 water, and I notice that an action takes place ; where the 
 light has not acted the gelatine is swelling up, showing 
 that it is absorbing water ; in other words, I see that the 
 lines forming the image are depressed, and the gelatine 
 around is in relief. I pour a gentle stream of water over the 
 surface, and then I wash away the soluble parts by the applica- 
 tion of a sponge. The lines are perfectly distinct, appearing 
 black on a white ground. Both of these properties of chro- 
 mated gelatine, which I have shown you, are utilised in what 
 we call photolithography. The images formed in the greasy 
 ink can be transferred to a lithographic stone and impressions 
 taken in the ordinary manner. This last process is known as 
 the Southampton method for preparing a photographic transfer 
 for lithography. The first method I showed you is certainly 
 equally as effective. It is also on these two properties of gela- 
 
32 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 tine, when in contact with the dichromates viz., non- 
 absorption of water and insolubility where light has acted 
 that a variety of other photographic printing processes are 
 
 
 Fm. 9. 
 
 founded. On the later reaction is founded the autotype process, 
 where the image is formed absolutely of coloured gelatine, all 
 the parts not acted upon by light having dissolved away ; 
 whilst on the former are built up all those processes which 
 produce prints in graduated tints of greasy ink after an image 
 has been obtained on gelatine that has been hardened and 
 rendered insoluble in water (though still leaving it capable of 
 absorbing water in the parts not acted upon by light), by the 
 addition of such substances as tannin, chrome alum, &c. As 
 examples of such processes I may mention the heliotype, 
 albertype, and autotype mechanical processes. 
 
 I have now come to the end of the time allotted to me, 
 and I trust that the explanations as far as they have gone 
 have been clear ; but in treating of such a large subject as 
 photography, it would be necessary for you to listen to me 
 for as many days as you have hours, in order that I might 
 enter into the details of much which I have merely been 
 able to glance at. 
 
THE ABSORPTION OF, LIGHT AND THE 
 COLOURS OF NATURAL BODIES. 
 
 TWO LECTURES. 
 
 BY PROF. STOKES. 
 
 LECTURE I. 
 
 THIS subject is one which does not admit very well of experi- 
 mental illustration before a large class ; in fact, with all the 
 appliances of the electric light, I should only be able to show 
 you, comparatively imperfectly, what you can each see for 
 yourselves by experiments which you can make quietly in 
 your own chambers, requiring, I may say, hardly any apparatus 
 at all. The foundation of what I have to say rests on Newton's 
 discovery of the compound nature of white light, with which 
 I presume you are already familiar. You know that when a 
 beam of light is allowed to fall upon a prism, it is decom- 
 posed into the different kinds of light of which it consists 
 which are bent round in passing through the prism to a 
 different degree. 
 
 Supposing a beam of sunlight reflected horizontally into a 
 room through a small hole and allowed to fall on a prism close 
 by, if the light were of one kind, the beam would be simply 
 bent round as shown in this diagram [referred to], and instead 
 of a circular spot being painted on the wall as at A, it would 
 be as at B. But on making the experiment you have actually 
 an elongated coloured image. The cause of that is, the light 
 is not of one kind, but consists of a variety of kinds differing 
 from one another by the colour with which they impress 
 the eye, and by their ref rangibility or capability of being bent 
 
34 LECTURES TO SCIENCE TEACHERS. 
 
 round in passing through, a prism, the red rays being bent 
 round the least, and the violet r&ys the most, while there are 
 kinds of light of all shades of refrangibility between the two 
 extremes. If I were to form a coloured image or spectrum 
 in this simple way it would not be what is called a pure 
 spectrum. Suppose, for simplicity of explanation, we had only 
 two kinds of light, blue and red, differing from one another 
 in refrangibility, then the incident light would be decomposed 
 into those two beams which would each diverge separately 
 from the source of light, or rather from the virtual images of 
 that source, and would be bent round to a very different 
 extent in passing through the prism; consequently if we 
 received them on a screen we should get two circular 
 patches of light, one blue and one red. Now actually, as 
 I have said, you have all intermediate shades of refrangi- 
 bility, and therefore this compound fan-shaped beam, which 
 passes through the prism, must be regarded as made up of 
 a vast number of cones of light differing from each other in 
 refrangibility, which increases from the red end to the blue 
 end. Consequently if you were to receive the whole on a 
 screen, any one point of the screen would not be illuminated 
 solely by one kind of light, but by all the kinds the refrangi- 
 bility of which lay within certain limits ; in fact there would 
 be a spectrum made up in this way, each circle that we draw 
 representing the section of one of those cones, and each over- 
 lapping the neighbouring circles. How then shall we arrange 
 to procure a pure spectrum, and that without loss of light ? I 
 say without loss of light, because a very simple mode, in 
 theory, of obtaining a pure spectrum would be to limit a beam 
 of light by one hole, and then by another at a distance ; the 
 diverging beam, which passes through the first hole, would be 
 limited by the second, so as to transmit only a very narrow 
 pencil of light, which you might regard as a mere ray, and if 
 you allowed that to fall upon a prism, it would be bent round 
 differently for the different kinds of light of which it consists, 
 so that you would get in that way a pure spectrum, but at an 
 enormous sacrifice of light. How then are we to obtain such a 
 pure spectrum without loss of light 1 This diagram represents 
 (Fig. A) a beam of sunlight diverging through a small hole, 
 forming a pencil of light. If that were received on a convex 
 lens at a sufficient distance (Fig. B), it would be brought to a 
 focus again on the other side, and would diverge from that 
 
ABSORPTION OF LIGHT, ETC. 
 
 35 
 
 focus afterwards. If we were to take a prism alone, place the 
 prism at a distance from the hole, and in its position of 
 minimum deviation (Fig. C) we should get, if there were two 
 kinds of light only, blue and red, two beams emerging in 
 different directions, the blue (represented in the figure by 
 interrupted lines) being bent round more than the red, 
 and diverging as if they came from two separate points. 
 Now suppose we combine these* two pieces of apparatus 
 
 . B. 
 
 together, placing the prism at a distance from the hole, and 
 the lens near the prism (Fig. D). Then the prism and the 
 lens each fulfils its own office, the prism causes each conical 
 beam of light to be bent round,, but differently, according to 
 the nature of the light, more for the blue than the red ; the 
 lens alone collects each of these conical beams, and brings it 
 ngain to a focus, and so this figure represents what will take 
 place. 
 
 D 2 
 
36 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 If a screen were placed exactly in the focus, and white 
 light containing light of all shades of refrangibility were 
 allowed to fall on the prisrn, each point of the screen whicli 
 was illuminated at all would be illuminated by one kind of 
 light only ; the different kinds would be separated from one 
 another on the screen, and consequently you would get a pure 
 spectrum, and that without the tremendous loss of light 
 encountered if you employ two holes placed at a distance 
 
 FIG. C. 
 
 from each other. The spectrum thus formed, though pure, 
 would be infinitely narrow ; but in order to give it breadth 
 you have only to substitute a line of light (in a direction 
 parallel to the edge of the prism) from the point of light ; in 
 other words, to transmit the light in the first instance through 
 a narrow slit instead of a small hole. This is the way in 
 which a pure spectrum is generally formed as a matter of 
 principle, though sometimes a mirror is used instead of a lens ; 
 
ABSORPTION OF LIGHT, ETC. 
 
 37 
 
 but I will not go further into that, because what I have said 
 is merely an introduction to the use of the prism in the 
 simplest manner possible. 
 
 The simplest way to form a pure spectrum experimentally, 
 and a way which suffices, provided you do not want to put 
 objects in the spectrum, but only to see it, is to suppose the 
 
 FIG. D. 
 
 lens to represent your eye, and the screen placed in the focus 
 of the lens to represent the retina. 
 
 I said that if the light was passed through a hole it would 
 be brought to a point ; but if it came in through a slit, which 
 you may regard as a succession of holes in a direction perpen- 
 dicular to the plane of the paper, then after passing through 
 the lens it would be brought to a line of light standing out 
 in a direction perpendicular to the plane of the paper, which 
 is supposed to be the plane of refraction for light from the 
 
38 LECTURES TO SCIENCE TEACH EE^. 
 
 middle of the hole, a line of blue light for the blue light, a 
 line of red light for the red light, &c., in different positions, 
 so that the spectrum on the screen would be made up of a 
 series of lines, red, yellow, and blue, &c., if there were so 
 many different kinds of light. Every pure spectrum from a 
 source of light allowed to fall through a slit is to be thought 
 of as made up of a number, generally an infinite number, of 
 images of the . slit, corresponding each to the light of one 
 definite refrangibility. Instead of having distinct images of 
 the slit, as you would if there were only a definite number 
 of degrees of refrangibility, if you have all shades you must 
 regard the spectrum as made up of an infinite number of 
 images of the slit placed side by side, and running one into 
 the other with no line of demarcation between them, except 
 in certain conditions in which there is a failure of certain 
 kinds of light. Now the simplest way of seeing this is to 
 take a slit, which may be of the roughest description, and a 
 prism which is just large enough to cover comfortably the 
 pupil of the eye, and to look at the slit through the prism as 
 I am doing now. I now see a coloured image or spectrum, and 
 in it the fixed lines of Fraunhofer. T presume you have heard 
 of these already, and I shall not describe them, as it would 
 take me too far from my subject. There is one point which I 
 must notice in the use of the prism held in this manner before 
 the naked eye. If you turn it round its axis, you will find 
 that for a certain azimuth of the prism the fixed lines of the 
 spectrum, or at least of a particular part that you are looking 
 at, will be seen distinctly ; but if you turn the prism a little this 
 way or that round a line parallel to its edge, they will become 
 indistinct. The particular direction or azimuth in which the 
 prism must be held is found by trial. You can focus the 
 spectrum by turning the prism one way or the other, until 
 the image you are looking at is sharp and clear; just as 
 in ordinary cases of focusing. The reason of that is easily 
 explained by geometrical optics, or the science which treats 
 of the mathematical consequences of the laws of reflection and 
 refraction of rays of light. It depends on the alteration of 
 the distance of the virtual focus from which the rays after 
 refraction through the prism come according to the azimuth 
 of the prism. I say virtual focus, although in point of fact, 
 after passing in that manner through a prism, unless it be in 
 the position of minimum deviation, the light diverges, not from 
 
ABSORPTION OF LIGHT, ETC. 39 
 
 a focus, but from two focal lines, as they are called. Now if 
 you combine the prism with a lens in order to project the 
 image on a screen, and allow homogeneous light to pass 
 through a hole at a sufficient distance, and to fall on the prism 
 when not in the position of minimum deviation, you will find 
 in one position of the screen a vertical line of light, and in 
 another position a horizontal line, and between the two you 
 will get a circular patch. Now suppose that the screen is 
 held in such a position that on it is formed the vertical line, 
 how is the image of a slit which you substitute for the hole 
 formed 1 It is formed by a succession of lines overlapping one 
 another in the direction of their length, which gives you in 
 fact a single straight line ; so that when white light is used 
 light of any one kind will be brought to a line on the screen, 
 or in this case on the retina of the eye, and the spectrum 
 will be seen distinctly. The particular azimuth in which the 
 prism must be held to see the spectrum distinctly depends on 
 the distance at which the slit is held from the observer, and 
 on the length of sight of the observer, and it will be different 
 from one end of the spectrum to the other. If I hold a prism 
 so as to see the red end distinctly, I must turn it a little to 
 see the violet end distinctly. Turning it in one way brings 
 the virtual image nearer to the eye, and turning it the other 
 way moves the image further off. If it is focused for the 
 red by holding the prism in a certain position, we must turn 
 it a little so as to diminish the angle of incidence, in order to 
 get the true focus for the violet. The reason for that is, that 
 there is no provision for chromatic compensation in the eye. 
 The eye in that respect is to be compared, not to an achro- 
 matic object-glass, but to a simple lens. The effect of the 
 dispersion of light as regards ordinary vision is not perceived 
 under ordinary circumstances, but it becomes very perceptible 
 indeed when you supply the eye with homogeneous light 
 of different kinds, or with light from which certain portions 
 are abstracted. I have tried the experiment of throwing a 
 pure spectrum on a printed page. On holding the page at 
 the usual distance of distinct vision, I was able to see quite 
 distinctly in the green and brighter part of the spectrum ; in 
 the red end I saw somewhat indistinctly from long-sightedness ; 
 and in the violet end very indistinctly from short-sightedness. 
 In choosing a prism it is very easy to see whether the glass 
 of which it is composed is good or not. If you look at the 
 
40 LECTURES TO SCIENCE TEACHERS. 
 
 prismatic image of a candle, and then, keeping that in the 
 field, move off the prism to arm's length, so as to get distinct 
 vision of the prism itself, by moving it about a little you 
 will be able to see whether it is free from veins or not. It 
 should be free from veins, although a prism for merely eye- 
 work need not be of the same excellence as if it were to be 
 used with a telescope. 
 
 . The more immediate object of my lecture is the coloration 
 of natural objects, and that is best studied in the first instance 
 in the case of clear coloured bodies such as solutions, or 
 coloured glasses. Here is a coloured solution, and if I reflect 
 the skylight through it you will see the colour is blue ; but 
 if I add a little more colouring matter to it, it is no longer 
 blue but red. The same effect exactly would be produced if, 
 instead of increasing the quantity of coloured fluid which was 
 mixed with the water, I had increased the thickness of the 
 stratum through which you looked in fact one is found by 
 experiment to have exactly the same effect as the other. What 
 is the colour of this fluid 1 If you saw it only in one stage 
 you would say it was blue, and if only in another you 
 would say it was red. It passes in fact from blue to red. 
 "What is the cause of that 1 You must remember that this fluid 
 is illuminated by white light, and white light is not all of the 
 same kind, but is a mixture of portions of light, differing from 
 one another by their refrangibility, and at the same time 
 differing from one another in the coloured impression which 
 they produce upon the eye. In glisses the colouring effect 
 upon the eye simply results from the super-position of various 
 kinds of light which are present. In order to study this 
 phenomenon and the cause of it, we must in the first instance 
 consider what would take place as regards one kind of .light 
 alone. If I had one kind of light (and approximately I 
 should get that by a Bunsen flame with a bead of common 
 salt introduced into it), supposing I viewed this through a 
 wedge-shaped vessel which I can slide in front of my 
 eye, if I begin where the thickness is nothing, no effect is 
 produced. In this particular fluid, if I had such a flame before 
 me, I should see at first no effect, and then, as I slid the vessel 
 so as to increase the thickness of fluid looked through, the 
 flame would become weaker and weaker, until I should not 
 be able to see it. The effect is one of a progressive weaken- 
 ing. The longer the path, of the light within this coloured 
 
ABSORPTION OF LIGHT, ETC. 41 
 
 or absorbing medium, the less the quantity of light which 
 escapes ; and the law according to which the intensity de- 
 creases is very readily obtained by a simple ^ consideration. 
 Suppose first we had a stratum of the fluid of a certain 
 thickness, say one-tenth of an inch, and this produced a 
 certain weakening in the light, or let through a certain per- 
 centage. Now if you treated the light which came through 
 to a second stratum, also one-tenth of an inch in thickness, 
 of the same fluid, it would let through the same percentage 
 as before, and so on. When you have to deal with a mass 
 of fluid you may in imagination divide it into strata each of 
 the same thickness. Suppose here is the horizontal surface of 
 a mass of coloured fluid, which we are observing in a vertical 
 direction with white light. The effect of the fluid on light of 
 any particular kind is simply to weaken it. If we divide the 
 fluid into strata of equal thickness, in passing through the 
 first the light is weakened in a certain proportion, depending 
 on the thickness of the stratum. In passing through the 
 second stratum the same percentage of the light will be let 
 through, and so on; so that in passing from stratum to 
 stratum the intensity of the light goes on decreasing in 
 geometric proportion. That is to say, each term of the 
 series expressing the residue bears to the preceding term of 
 the series the same ratio throughout. Consequently when 
 you get far enough into the stratum the light has been so 
 much weakened that it becomes altogether invisible. Theore- 
 tically, however great the stratum, there is a quantity which 
 still gets through, but practically, after a certain time, the 
 quantity which gets through is so very small that it may be 
 regarded as nothing at all, and the light is extinguished. 
 Now the rate at which the light is so extinguished depends 
 upon the kind of light which falls upon the coloured stratum. 
 Suppose that rate to be different for different kinds of light, 
 then if there were only two kinds presented to the fluid in 
 the first instance, as it passed on and on, through this ab- 
 sorbing medium, the proportion of these two kinds of light 
 would be continually changing. For the sake of clearness I 
 will suppose there are two kinds of light to start with, blue 
 and red, and that at the beginning blue has an intensity of 
 100 and the red of 10. There is of course a great predomi- 
 nance of blue over red. Now suppose in passing through a 
 stratum of a certain thickness half the blue light is lost, and 
 
42 LECTURES TO SCIENCE TEACHERS. ' 
 
 only half transmitted, and that ninety per cent, of the red 
 light is transmitted. Then after passing through the first 
 stratum the intensities will be respectively 50 and 9 ; after 
 passing the second stratum of the same thickness the intensities 
 will be 25 and 8'1 ; after the third 12'5 and 7'3 ; after the 
 fourth 6 - 2 and 6 '6, or about equal; but afcer passing through 
 the next stratum they will be 3'1 and 5'8 ; so that 
 although the quantity of red light was so much smaller lo 
 begin with, the red is more lasting, and in light which has 
 passed through five of these strata the red now predominates 
 over the blue. Passing through another stratum, the in- 
 tensity of the blue is reduced to 1*5, whilst the red is 5*2, 
 and so on ; so that you see both kinds of light are weakened 
 but the proportion to one another is continually changing. 
 That is a general explanation of what takes place in a fluid 
 such as this [exhibiting an alkaline solution of archil]. I 
 may mention that in almost all coloured fluids there is a con- 
 tinual change in the colour according to the thickness of 
 the stratum of liquid, or, which will come to the same thing, 
 according to the strength of the solution. For the sake of 
 simplicity of explanation, I supposed there were only two 
 kinds of light to deal with, which I called red and blue, but 
 in point of fact when the fluid has white light thrown upon 
 it we have an infinite number of kinds of light, and all shades 
 of refrangibility, and each shade of refrangibility must be 
 considered by itself. If we take a certain stratum of a 
 coloured liquid or glass, or whatever it is, then after passing 
 through that stratum the light will be weakened in a pro- 
 portion which changes continuously in passing from one end 
 of the spectrum to the other. The mode in which the total 
 light passing through the stratum is made up may very con- 
 veniently be represented to the eye by a construction given 
 by Sir John Herschel in his treatise on light. Suppose you 
 take a horizontal line and lay distances along that line, or 
 abscissae, representing the places of the kinds of light which 
 you have under consideration in some standard spectrum, and 
 let lines drawn vertically, or ordinates, represent the intensity 
 of the particular kind of light. If you care merely to know 
 how the quantities go on changing as the light passes deeper 
 and deeper into the absorbing fluid or glass, it will be 
 simplest, to take the original intensity as unity throughout, 
 although we know very well that the different parts of the 
 
ABSORPTION OF LIGHT, ETC. 43 
 
 spectrum are not equally bright. But that is a point which 
 we need not for the moment take into consideration. A 
 horizontal line, then, parallel to the axis of abscissae, may be 
 supposed to represent for each particular colour, the place 
 of 'which is defined by the abscissa, the original intensity of 
 that colour. Now after passing through a certain stratum of 
 the medium of a certain thickness, that intensity will be re- 
 duced differently for the different colours, and consequently 
 the locus which defines to the eye the composition of the 
 light which is passed through that stratum will be a certain 
 curve, but it will depend on the nature of the medium 
 what the nature of the curve will be. Now I have drawn 
 here [referring to figure] what represents a curve for a green 
 colour in a certain medium. To find how the light will be 
 composed after passing through a second stratum of equal 
 thickness, we have nothing to do but for a sufficient number 
 of abscissae to take an ordinate which bears to the ordinate 
 of this curve the same ratio that the latter bears to the 
 original ordinate or unity, and so on for additional strata 
 of the same thickness. Thus we get a succession of. curves 
 representing to the eye the composition of the light which 
 has passed through successive thicknesses of the medium. 
 You may notice that the quantity of light altogether goes 
 on decreasing ; but that is not all, the proportion of the 
 different parts goes on changing as well. In this case, if 
 the opacity of the medium is such as is represented in this 
 curve, the blue or bluish-green light which predominated a 
 little at first will predominate more and more, and the 
 colour of the medium will become a purer and purer green 
 as you look through greater and greater thicknesses of it. 
 Here is another curious curve representing in the same 
 manner the type of light which is transmitted through one 
 of the ordinary blue glasses coloured by oxide of cobalt. I 
 do not pretend that it is an exact representation, but it is 
 an approximate one, and you will see the curious alternations 
 which there are in this case, of comparative opacity and 
 transparency. In this way we can readily understand how 
 it is that the colour of a coloured fluid is continually 
 changing according to the thickness looked through. This 
 phenomenon in its more striking examples is sometimes 
 called dichroism, but as that word has been employed to 
 designate so many phenomena totally different from one 
 
44 LECTURES TO SCIENCE TEACHERS. 
 
 another in their mode of production, I hardly like to em- 
 ploy it. 
 
 You have seen in what manner you can readily, and 
 almost without any apparatus, observe a pure spectrum, and 
 how it is modified by the interposition of a coloured body ; 
 and I may just mention one or two instances of interesting 
 results which may be obtained in this manner. Sometimes 
 the mode in which a coloured medium attacks the different 
 parts of the spectrum is highly characteristic of the particular 
 iluid that you are employing. Here, for example, is one very 
 characteristic case the red colouring matter of blood. The 
 spectrum which that gives is represented in the upper part of 
 this figure [referred to]. In order to see the spectrum nothing 
 more is requisite than this : You take a slit of the roughest 
 description here is one made of wood and tinned iron 
 blackened, and the blood is conveniently held in a test-tube, 
 which you can hold in position by an elastic band. In order 
 to see the spectrum by transmission you have nothing more to 
 do than to hold this against a source of light and look at it. 
 If you use, not a wedge-shaped vessel, but a test-tube, you 
 cannot be sure of not passing over some of the most interesting 
 parts of the phenomena, unless you go step by step, and use 
 several different thicknesses, or, which comes to the same 
 thing, different degrees of dilution. For instance, when a 
 solution of blood is so highly coloured as this, a great part of 
 the spectrum is cut off, and it may be that you will see 
 nothing but a broad black band, whereas, if I had used a 
 weaker solution or a test-tube of smaller diameter, I should have 
 seen certain highly characteristic phenomena of absorption. 
 In order to see these, the solution must be so diluted that it 
 is little more than pink. Then you will see these highly 
 characteristic dark bands of absorption. I know of no sub- 
 stance which can be confounded with blood if you simply 
 take the spectrum of it in this manner, unless possibly an out- 
 of-the-way substance, turacine, the colouring matter found in 
 the red feathers of the wings of the touraco, a bird found at 
 the Cape of Good Hope. If you only looked at the spectrum 
 in one condition, it is possible that the two might be con- 
 founded, although hardly so ; but if you combine the obser- 
 vation of one of these peculiar spectra with the observation 
 of the effect of re-agents, you get a combination of characters 
 which is such that it is almost impossible to confound any 
 
ABSORPTION OF LIGHT, ETC. 45 
 
 other substance with the one which you have under your 
 hands. This becomes a mode of discrimination between sub- 
 stances of the utmost value to chemists, but which, strangely, 
 for a long time they altogether neglected, though, since the 
 researches of Kirchhoff and Bunsen, the chemical spectroscope 
 has become an instrument in the hands of almost every 
 chemist. 
 
 I may mention one reaction with reference to the colouring 
 matter of blood which is interesting in itself, and will also 
 illustrate what I am saying. You know that the venous 
 and arterial blood differ from each other in colour. If you 
 look at the veins at the wrist you can see the redness of 
 the arterial blood in the arteries which happen to be near 
 enough to the surface, as contrasted with the deeper and darker 
 colour of the venous blood. This difference can be imitated 
 by introducing into a solution of blood a suitable deoxidising 
 agent, which will alter its colour. I have here, in the first 
 instance a solution of protosulphate of iron, and I have added 
 to that tartaric acid, which has the property of preventing the 
 precipitation of many metallic oxides. The colouring matter 
 of blood is immediately decomposed by acid, and therefore 
 you must take care not to introduce acid into the solution. I 
 have rendered this solution alkaline by ammonia without pre- 
 cipitating the iron. This is a strong reducing agent. It is 
 in small quantity almost colourless, and if a little of that is 
 introduced into the colouring matter of blood, which is 
 not decomposed in any reasonable time by ammonia, then 
 immediately the colour is changed into a purple one, and the 
 spectrum is changed in a remarkable manner, as represented 
 in the lower half of the diagram. In lieu of the two dark 
 bands, you have a single band occupying an intermediate 
 position. The fluid is purpler than before, and lets through 
 more blue light. If you have such a solution in a test-tube, 
 and shake it up with air so as to re-oxidise it, you get back 
 the original solution, and you may put it backwards and 
 forwards as often as you like. But I merely mention this as 
 illustrating what you get by using simply a prism without 
 any apparatus at all, and you can see the actual spectrum as 
 shown on the diagram. If a test-tube containing a solution 
 of blood deoxidised in this manner be allowed to stand for 
 some time, it absorbs oxygen from the air, the upper part 
 becomes oxidised, and this oxidation extends deeper and 
 
46 LECTURES TO SCIENCE TEACHERS. 
 
 deeper down, and after a certain time the tipper portion of 
 the blood is seen of the scarlet colour, and the under portion 
 of the purple colour. If you then put the test-tube behind 
 a slit, such as I have shown you, and look at it through a 
 prism, you will see the two spectra simultaneously, as repre- 
 sented in the figure. 
 
 To take another example, I have here a solution of per- 
 manganate of potash. If it is considerably diluted, and you 
 analyse the light transmitted through, you will see a broad 
 dark band in the spectrum. If you have it more diluted, you 
 obtain a spectrum highly characteristic, in which are seen five 
 dark bands in the green part of the spectrum. Those are 
 highly characteristic of the permanganates. There is just a 
 trace of a sixth band, which comes in when the solution is 
 stronger. These are alternations of transparency and opacity, 
 not that the fluid is perfectly transparent, but these inter- 
 vening spaces are really alternations of greater and less absorp- 
 tion. When the quantity present is sufficient, the whole of 
 this region is absorbed, and then the characteristics are lost, 
 because there are a great variety of purple substances which 
 would give a spectrum not very different. In examining a 
 substance you must dilute the solution to make sure of break- 
 ing up any such broad dark region, and then you see the 
 dark bands, if any, which are characteristic of the substance. 
 There are other red solutions of manganese which may be 
 obtained, and which agree with the permanganates in being 
 powerfully oxidising agents, and which long ago were con- 
 founded by chemists with the permanganates because they 
 have both- the purple colour, and are powerfully oxidising 
 agents. For example, if you rub up binoxide of manganese 
 with binoxalate of potash, you obtain one of these purple 
 coloured solutions, though it is not very permanent, which as 
 being a powerful oxidising agent and also of a purple colour, 
 was supposed to contain permanganic acid, but the spec- 
 trum instantly shows you it is nothing of the kind. These 
 two examples will suffice to show how valuable the prism is, 
 even without any other apparatus, as a means of discrimi- 
 nating between different bodies. 
 
 The phenomena of the coloration of natural bodies is 
 best studied, as I said before, in coloured solutions ; but I now 
 pass on from that to the colours of natural bodies as commonly 
 presented to us. Let us take, for example, a very common 
 
ABSORPTION OF LIGHT, ETC. 47 
 
 colour, the green of vegetation, as in grass and leaves in 
 general. What is the cause why a green leaf is green, or 
 why a red poppy is red? It is frequently said that the 
 reason why a red poppy is red and that a white lily is 
 white is, that the lily reflects rays of .all kinds, but the poppy 
 reflects only the red ones, and if you place the red poppy in 
 a pure spectrum it is luminous, like a white lily, in the red ; 
 hut if you place it in the green it will be almost black, 
 whereas the white lily will be brilliantly green. JSTow the 
 common explanation, properly understood, is true ; tjut it is 
 not the whole truth, and if understood as it is liable to be 
 understood, it is false. It is true that a red poppy reflects 
 red rays, and a white lily reflects rays of all colours ; but it is 
 not true that the preference for the red to the green in the 
 one case and the equality of action in the other takes place 
 in the act of reflection. It is not a phenomenon of colora- 
 tion by reflection. The coloured light is reflected, or you 
 would not see it ; it is sent out of its course before it 
 enters your eye, and it is true that the light, in its life's 
 history, undergoes reflection ; it is not true that it is in the 
 act of reflection that the one colour gets the preference over 
 the other. Here I have some solution of the colouring 
 matter of green leaves in alcohol, and here is some more 
 alcohol, with which I will dilute the former. I have obtained 
 a beautiful green solution, although the green colour is not 
 seen now by reflected, but by transmitted light. As regards 
 the light which falls upon the surface, there is a little white 
 light reflected, just as there would be from water, but very 
 little is reflected from the surface where the fluid is in contact 
 with the glass ; the chief portion of that reflected being from 
 the outer surface of the glass itself. You would not see any 
 green at all in it unless there were something placed behind 
 so as to reflect the light backwards. You see there that the 
 colour of the green leaf, as ordinarily seen, is due to the 
 combination of reflection with the phenomena of absorption, 
 or the swallowing up of certain kinds of light when light is 
 sent through a perfectly clear medium. I may illustrate this 
 in another manner. Here is a vessel of water, into which I 
 will pour some blue solution. If I send light through it, it 
 will appear of a deep blue, but if I hinder the light from 
 coming behind, which I can do by putting black cloth behind 
 it, it is simply dark ; you do not see the blue colour at all. 
 
48 LECTURES TO SCIENCE TE AGREES. 
 
 Why 1 Because there is nothing behind to reflect the light. 
 Suppose I make it a little muddy by pouring into it some 
 pounded chalk, you see the blue colour immediately. Why 
 is that 1 ? You know that if powdered chalk were put into 
 water it would not colour the fluid. J>ut here each little 
 particle of uncoloured chalk reflects a small quantity of .light 
 falling upon it, so that it fulfils the same office as a mirror 
 placed behind the fluid. You may imagine that the particles 
 of chalk are so many minute mirrors capable of reflecting 
 light. If you take any one particle of chalk, say one-tenth of 
 an inch deep, in the liquid, the light from the sky falls upon 
 the fluid, it undergoes absorption in passing through that 
 first tenth of an inch, and then the portion of light which is 
 left is reflected by that little particle of chalk, and passes out. 
 again, and so, as regards that single particle, the light which 
 reaches your eye from beneath that depth has itself gone 
 through a stratum of fluid of one-fifth of an inch in thickness, 
 and accordingly you see the colours produced by selective ab- 
 sorption, that is to say, by the absorption of certain kinds of 
 light, which are more greedily devoured by the fluid than the 
 other kinds. This is what takes place in the green leaf, and 
 in the petals of flowers. Let us take the white lily. If the 
 petal of the flower had been merely a sheet of thin glass, you 
 would not have seen that white colour. There would have 
 been a little light reflected from the first surface and the back 
 surface, but the petal is really composed of a vast assemblage 
 of little cells, at each of which partial reflection takes place, 
 so that it resembles some finely-powdered glass, which would 
 form a white powder, because each little surface is capable of 
 reflecting the light, although a single sheet of glass would not 
 be white. The petal of the white lily is just in the condition 
 of the powder. It is full of little cells, full, optically speak- 
 ing, of irregularities, from each of which a portion of light 
 is reflected, so that, all kinds being reflected alike, and there 
 being nothing in the white lily to cause preferential selection 
 of "one over the other nothing to sift the light, as it were 
 you get a considerable quantity of light reflected back to 
 the eye, bat it is white. What is the difference between that 
 and the red poppy 1 The red poppy is, as it were, a white lily 
 infused with a red fluid ; there is light continually reflected 
 backwards and forwards, just as before, at the surface of the 
 cells ; but that light, in going and coming, passes through the 
 
ABSORPTION OF LIGHT, ETC. 49 
 
 coloured juice of the plant. It is the same thing with a green 
 leaf. The structure is irregular, optically considered ; there are 
 constantly reflections, backwards and forwards, of light, which 
 penetrates a little depth and is reflected, and has to pass 
 through a certain stratum of this colouring matter, to which 
 the name chlorophyll has been given, but which is really a 
 mixture. That is what takes place generally as regards the 
 coloration of bodies ; it is a phenomenon not of reflection, 
 not of selection of one kind of light for more copious reflec- 
 tion than anqther, but of absorption, or the swallowing up of 
 certain kinds of light. Keflection comes in, in order to enable 
 us to see the light which otherwise would not enter the eye 
 at all, but would go off in another direction. 
 
 The spectrum of this green fluid, which is a , substance to 
 which I have paid a great deal of attention, is very peculiar. 
 It is a mixture of several substances with closely-allied 
 chemical properties. The peculiar spectrum may be seen in a 
 green leaf itself, if you place it behind a slit and analyse it by 
 transmitted light, or allow a strong light, such as that of 
 the sun, to fall upon it and analyse the reflected light. 
 
 Now you will say, Are there no colours in any case produced 
 by reflection 1 ? is there no case in which this preferential 
 selection is made 1 How is it, for instance, if you take a plate 
 of gold ; that reflects light regularly, but the light is coloured 
 yellow 1 I said the cause of the coloration of bodies in the 
 great bulk of cases was what I have just described, but I did 
 not say that was the sole cause of coloration. The light 
 reflected from gold is in fact coloured ; in the case of gold or 
 of copper there is a preferential selection in the act of reflec- 
 tion of one kind of light rather than another, and that 
 preferential selection is not confined to the metals, although 
 it is chiefly in gold and copper that we ordinarily perceive it. 
 There are many cases in which substances which absorb light 
 with intense avidity present a similar reflection of coloured 
 light, and in these substances the connection between the 
 intense opacity of the substance and the coloured reflection 
 can be better studied than in the case of metals, because a 
 metal is, under ordinary circumstances, opaque. Certain of the 
 aniline colours, for instance, show the coloured reflection in a 
 notable manner. These specimens on the table (referring to 
 plates of glass on which solutions of certain aniline colouring 
 matters had been evaporated) were given to me by the late 
 
 
 
50 LECTURES TO SCIENCE TEACHEES. 
 
 Sir Charles Wheatstone. He prepared them himself. This 
 is a deep blue or purple by transmitted light, but it is an 
 exceedingly thin film, and by reflected light it has a bronzy ap- 
 pearance. Here is another which is green by reflected light 
 and red by transmitted light. In these cases we see that we 
 have a substance which does exercise a preferential selection 
 for one kind of light as compared with another in the act of 
 reflection. But the light which is so selected for preferential 
 reflection is not at all the light which is chiefly transmitted ; 
 on the contrary, it is the very reverse. If we analyse the 
 light transmitted through this red stratum, or through a solution, 
 we shall find that in the green part of the spectrum the 
 substance is more intensely opaque than elsewhere ; that is 
 to say, the film must be excessively thin, or the solution exces- 
 sively dilute, in order that any light at all strong enough to 
 be seen may get through in the green part of the spectrum. 
 The substance is intensely opaque as regards the green, but 
 moderately opaque only as regards the other parts. A 
 solution of this colouring matter does not present this 
 coloured reflection at all. The colouring matter must be 
 excessively concentrated, as it is when a solution of it is dried 
 on glass, in order that this reflection should be shown, and 
 then the kind of light which is more especially reflected in 
 that manner agrees with the kind of light which is intensely 
 absorbed. Those parts of the spectrum which are absorbed 
 with this enormous intensity, so that the dry film is with 
 regard to them as opaque as a film of metal of the same 
 thickness would be, or thereabouts, are reflected as copiously 
 as they would be by a metal, and the colours which are only 
 moderately absorbed are reflected very much as they would 
 be by the glass, and accordingly in the reflected light there 
 is a predominance of those colours which are intensely 
 absorbed. The most remarkable example, that I know of, 
 of the connection between intense absorption and powerful 
 reflection, takes place in the case of crystals of permanganate 
 of potash. These crystals have a bronzy look by reflected 
 light when freshly taken out of the mother liquor, so that 
 the surface is not spoilt by tarnishing, as soon happens from 
 exposure to the atmosphere ; the sides of the crystals have a 
 metallic brilliancy, and reflect green light. Now that light 
 agrees with the light reflected from a metal, not only in 
 its copiousness, but also in certain other properties. Jf I 
 
ABSORPTION OF LIGHT, ETC. 51 
 
 take light reflected from glass at a certain angle, which is 
 called the angle of polarisation, the reflected light is polarised, 
 and capable of being extinguished by an analyser such as a 
 Nicol's prism. Light reflected from a metal is not polarised at 
 any angle of incidence, though it is partially polarised at an 
 oblique angle. I say partially polarised, but I will leave the 
 explanation of that to my friend Mr. Spottiswoode, who will 
 give a lecture on that subject. Bronzy crystals of perman- 
 ganate of potash agree in that respect, to a certain extent at 
 least, with the metals ; if you examine the light by reflection 
 you find that it is not capable of extinction by analysing 
 under any conditions. If you examine it at such an angle 
 of incidence that a vitreous substance would give you light 
 capable of extinction, the light becomes weaker and of purer 
 green. I have analysed the' light reflected by a crystal under 
 these conditions, by a combination of a prism and a Mcol's 
 prism, so as to extinguish what light would have been reflected 
 from glass under similar conditions, and this curious result came 
 out. I must premise that crystals of permanganate of potash 
 are too intensely opaque to allow you to examine them by 
 transmission, but you can make a solution of them and ex- 
 amine that, and it shows these bands of absorption which are 
 shown on the diagram [referred to]. Now on examining, in 
 the manner I have mentioned, the green light reflected from 
 the crystals at an angle similar to that at which light reflected 
 from glass would have been quenched by a Nicol's prism, this 
 curious result was obtained ; the spectrum was seen to consist 
 of four bright bands, and perhaps a trace of another, the rest 
 of the spectrum being wanting. Now what were the positions 
 of those four bands ? When the positions were observed, by 
 referring them to the standard fixed lines of the spectrum, 
 which were seen at the same time, they were in the positions 
 represented in the under part of that figure ; they agreed in 
 position with the first four of the five dark bands seen in the 
 transmitted light. The spectrum begins to get comparatively 
 faint in the region of the fifth band of absorption, and there 
 was hardly a chance of seeing the fifth bright band if it had 
 been there ; but you see that whereas, as regards transmitted 
 li.^ht, the crystals pass alternately through maxima and 
 minima of transparency alternately from the . condition of a 
 vitreous substance to the condition of a metal, as to the avidity 
 with which they absorb the light corresponding to these 
 
 E 2 
 
62 LECTURES TO SCIENCE TEACHERS. 
 
 alternations you have also alternations in the character of 
 reflected light ; so that you may say the substance is alter- 
 nately opaque and transparent, comparatively speaking only, 
 as regards the transmitted light, and, corresponding to these 
 alternations, it behaves as regards reflection alternately as a 
 metal and as a vitreous substance. That shows how the 
 coloured reflection, where it does exist it is a phenomenon, 
 comparatively speaking, rare is connected with the quasi- 
 metallic opacity of the substance as regards transmission. 
 
 You may say that if that be the case the colour of gold 
 ought to be not yellow at all by transmission ; nor is it. Gold 
 leaf is thin enough to allow some light to pass through it 
 otherwise than by mere holes, which occur accidentally here 
 and there, and that transmitted light is green. I have here a 
 little chloride of gold in solution. I put a little protosul- 
 phate of iron in it, and if the experiment is properly per- 
 formed you obtain what is not really a solution of gold, but 
 gold suspended in a state of exceedingly fine division ; and 
 in that way, when the fluid is looked through, you get it 
 distinctly blue, which is the real transmission colour of gold. 
 I have seen the same thing with regard to copper. Dr. Percy 
 gave me a specimen of a very curious glass, which I intended 
 to have brought with me. The ordinary red glasses are 
 coloured by suboxide of copper, which is put over a piece of 
 colourless glass in a film of copper-salt so thin that you do not 
 see any colour at all by light transmitted directly across, 
 but where you look through obliquely you can just see the 
 faintest possible blueness. The film of copper-salt is reduced 
 by a suitable agent to a silicate of suboxide, which gives 
 that beautiful red colour, which is contained in a film thinner 
 than the thinnest paper. In this case the glass was covered 
 with copper in a similar manner, but it was a deep blue 
 by transmitted light, and if you play on any particular spot 
 with a blow-pipe it becomes sensibly colourless. The 
 colouring matter was copper, but in what state 1 Evidently 
 in this case the reduction necessary for reducing the oxide 
 of copper to suboxide had gone on rather too far, the 
 copper was reduced to the metallic state ; you looked through 
 the copper, and it was seen to be blue. So that you see 
 that in the same sense in which the coat of an English 
 soldier is red, the colour of gold is blue or green, and the 
 colour of copper is blue. There is the same relation there 
 
ABSORPTION OF LIGHT, ETC. 53 
 
 as in this aniline glass between reflection and absorption, 
 but whereas in the aniline colours it is commonly the 
 phenomena due to absorption, and the selection of one kind 
 of light over another in the act of transmission, which 
 meet your eye, in the case of the metals gold and copper 
 it is a selection which takes place in the act of reflection 
 which ordinarily presents itself to observation, and the true 
 colour by transmission is only seen under very exceptional 
 circumstances. 
 
FLUORESCENCE. 
 
 BY PROFESSOR STOKES. 
 
 LECTURE II. 
 
 THE subject which I am about to bring before you to-day 
 is one which has attracted a great deal of attention for some 
 years back, and in which 1 have myself had a considerable 
 share. One of the first phenomena discovered in connec- 
 tion with those I have to bring before you was that which Sir 
 David Brewster called the internal dispersion of light, which 
 he first noticed in an alcoholic solution of the green colouring 
 matter of leaves, as mentioned in a paper read before the 
 Pvoyal Society of Edinburgh in 1833, 1 and fully described 
 in a later paper read before the same Society in 1846. 2 I 
 have here a solution of the green colouring matter of leaves 
 which I used in my lecture on absorption, but which I am 
 now going to use for a different purpose. Brewster had 
 occasion to pass a beam of sunlight through this green fluid, 
 and he was surprised to observe the whole of the path of 
 the beam exhibiting a blood-red light. The figure which I 
 have here^ is intended to represent what could be seen, and 
 I will endeavour to show it presently. This represents a 
 vessel filled with the green fluid and placed on a white ground 
 such as paper, with a glass bottom so that you can see the light 
 through. In looking through you see the green colour of 
 the solution, and there is supposed to be a board standing 
 vertically on its edge containing a lens. The sun's light is 
 reflected horizontally and sent through that lens so as to 
 form a condensed beam, the focus of which lies within the 
 
 1 Edinburgh Transactions, xii. 541. 
 
 2 Ibid. xvi. Ill ; or Phil. Mag. for June, 1848. 
 
 3 The figures referred to in this lecture are not reproduced, except 
 in two cases, as they are mostly coloured, and would lose much of 
 their significance if merely represented by black and white. 
 
FLUORESCENCE. 55 
 
 vessel. When that is done, you see the whole path of the 
 beam marked by this blood-red light This is a -very curious 
 phenomenon : what is the cause of it ? Sir David Brewster 
 seemed to imagine that the ultimate particles of the 
 substance reflected red light somewhat in the manner of 
 finely suspended vermilion. Suppose in fact you had a 
 fluid which was green by transmitted light, and you could 
 manage to form in that an excessively fine mud of ver- 
 milion, then it is conceivable that you might get a pheno- 
 menon of this kind ; I do not say that is the true cause, for 
 it is not. Brewster examined a number of substances, both 
 solutions and solid bodies, in a similar manner, and I may 
 mention one which is described by him in a later paper read 
 before the British Association in 1$38, a certain variety of 
 fluorspar. One of the varieties he mentions is a green kind' 
 as seen by transmitted light, found at Alston Moor in Cum- 
 berland, and I may mention that there is another variety, 
 which usually is purplish by transmitted light, which 
 abounds in Mr. Beaumont's lead mines at Allenhead, which 
 shows the phenomenon even better. This kind of fluorspar 
 shows a deep blue light in certain aspects. You see that to 
 perfection, if you plunge the spar into water, because then 
 you get rid in a great measure of the light reflected from 
 the surface. When a condensed beam of sunlight is ad- 
 mitted into the crystal, the path of it is marked by a 
 blue light. It is not, however, continuous, like the red light 
 in the green fluid, but it occurs in strata parallel to the 
 nearest faces of the cube. Evidently it depends upon some- 
 thing which took place during the growth of the crystal. 
 Possibly it may have crystallized thousands of years ago, 
 we know not how long, out of a solution, the nature of 
 which gradually changed as the crystal grew, and some sub- 
 stance probably was taken up by the crystal, to which this 
 effect is due. 
 
 Some years later Sir John Herschel published a paper in 
 the Philos<>t>liical, Transactions, " On a case of superficial colour 
 in a colourless liquid," which was shortly afterwards followed 
 by a paper " On the epipolic dispersion of light. 1 ' Quinine as 
 you know is very much used in medicine and when a solu- 
 tion of quinine is formed, tolerably dilute, in water acidulated 
 
 1 Philosophical Transactions, Jan. 1845, pp. 143, 147. 
 
56 LECTUEES TO SCIENCE TE AGREES. 
 
 with sulphuric acid, by transmitted light the fluid looks very 
 much like water, but it exhibits in certain aspects a blue 
 colour. You will not perhaps very well see it here, but this 
 is such a common fluid, easily obtained by anyone, that it is 
 almost sufficient to mention the appearance. What is 
 remarkable about this blue colour is that (unless the solu- 
 tion be excessively dilute) it is mainly concentrated, and 
 occurs in an exceedingly narrow stratum adjacent to the 
 surface by which the light enters the fluid. This diagram 
 [referred to] represents the appearance. 
 
 This [referring to figure] is supposed to be a section of a 
 tumbler containing the solution, placed on a black ground, 
 and tolerably near a window from which light is coming in 
 approximately horizontally. "When you look down from 
 above you see that the side of the fluid next the window 
 is marked by this bluish colour, and when you hold the eye 
 almost in a prolongation of the anterior surface of the fluid 
 you see this blue stratum very much foreshortened and 
 thereby increased in intensity, because the fluid itself is 
 transparent like water, and the blue light which appears, 
 whatever may be its cause, is seen perfectly well through 
 it. When you look down in an oblique direction you see it 
 much less intense. It is seen in perfection on allowing the 
 light to shine from above, holding the eye a shade below the 
 level of the upper surface, and putting a black object to 
 make a dark background. 
 
 Now what is the nature of this blue light] Sir John 
 Herschel tried various experiments on it. He analysed it 
 by a prism, and obtained a continuous spectrum. He 
 noticed, however, that he did not see the Fraunhofer lines 
 in the spectrum ; but whether they were really absent, or 
 that he did not see them because the light was not strong 
 enough, he does not seem quite decided. He noticed also 
 that the blue light exhibited no trace of polarisation. He 
 examined further the light transmitted through the solution 
 to see what blue rays were taken out of it. Naturally he 
 was led to scrutinise more particularly the blue part of 
 the spectrum, but apparently the blue part of the spectrum 
 was like the blue part of the spectrum of light which had 
 come through simple water ; there was nothing particular 
 to be seen in it to account for the phenomenon. Possibly 
 ovever, if this superficial colour is produced once only, 
 
FLUORESCENCE. 57 
 
 the quantity of light removed from the spectrum may not be 
 sufficient to show any dark bands of absorption in the blue 
 region, but if you repeated the process on the light, making 
 it pass through different vessels in succession giving out 
 this blue stratum at the surface of each, perhaps then you 
 would have sufficiently weakened the blue of the transmitted 
 spectrum to show what particular rays were taken out by 
 the fluid. Sir John Herschel however observed that when the 
 light had passed through a thin stratum of the fluid in the 
 first instance, though it resembled ordinary light when it 
 came out, it had lost its power, for some reason or other, of 
 producing this phenomenon. What the power was did not 
 at the time further appear. Sir John Herschel called the 
 phenomenon ejripnlic dispersion from a Greek word signify- 
 ing surface, and he called the light which having passed 
 through a moderate thickness of solution of quinine had 
 been shorn of the power of producing that effect, epipolized. 
 In one of his experiments he had occasion to throw 
 sunlight vertically downwards on the fluid, and in that case, 
 the light being pretty strong, he observed the blue colour 
 extending to a depth of half an inch or more into the solution. 
 It was much stronger at the surface, but extended a con- 
 siderable way down. 
 
 After the appearance of Sir John Herschel's paper, Sir 
 David Brewster took up the subject and examined this 
 particular fluid, the solution of quinine, as he had done 
 the solution of the green colouring matter of leaves and 
 fluorspar, and various solutions, 1 and he found that when 
 a beam of sunlight, concentrated by a lens, was admitted into 
 a solution of the quinine in dilute sulphuric acid, the whole 
 of the path of the beam was marked by this blue light. At 
 the same time, if you repeat the experiment, you will see at 
 once that the blue colour is decidedly more copious in the 
 immediate neighbourhood of the first surface. He further 
 examined the beam as to its polarisation, by viewing it 
 through a rhomb of calcareous spar, and stated that a con- 
 siderable portion of it, consisting chiefly of the less refran- 
 gible of its rays, was polarised in the plane of reflection, 
 while the greater part, constituting an intensely blue beam, 
 was found to be unpolarised. It is almost impossible to get 
 
 1 See the paper already referred to, Edinburgh Transactions, xvi., or 
 Phil. Mao. June 1848. 
 
58 LECTURES TO SCIENCE TEACHERS. 
 
 a fluid or solution like this perfectly free from motes, and 
 the motes which are present will reflect a certain quantity 
 of white light, and that light is principally polarised when 
 looked down on vertically from above in a plane passing 
 through the beam. When the beam is viewed by light 
 polarised in the perpendicular plane, any light which would 
 be reflected from motes is nearly got rid of, and the blue 
 light is seen in its purity. 
 
 In this mode of observation it clearly appeared that the 
 solution of quinine belonged to the class of bodies in which 
 Sir David Brewster had discovered what he called internal 
 dispersion. Among them there is a kind of glass which 
 exhibits it in a very remarkable degree Here is a specimen 
 of the glass ; it is coloured by sesqui-oxide of uranium. He 
 noticed in this case, that the whole of the beam was un- 
 polarised, or, as he expressed it, possessed a quaquaversus 
 polarisation. 
 
 It was twenty five years ago last Easter, when I was pre- 
 paring for my optical lectures in Cambridge, that, having 
 had my attention directed by a friend to this solution, I 
 procured some, and I was greatly struck with the remarkable 
 appearance of the phenomenon, and the question occurred 
 to me what was the cause of it. Now for my own part I 
 had the fullest confidence in the doctrine that, the light 
 belonging to a given part of the spectrum is homogeneous, 
 or all of the same kind. I am not now speaking of polari- 
 sation. It has been supposed by some, by Sir David 
 Brewster, for instance, that the light of a given part of 
 the spectrum, although no longer decomposed by the prism, 
 might be decomposable by other means, for example by 
 the use of absorbing media; and he thought he had obtained 
 white light from a particular part of the spectrum by the 
 use of a suitable absorbing medium. This has since been 
 proved to be merely an illusion of contrast. Having, as I 
 said, felt the fullest confidence in the principle that the light 
 of a given part of the spectrum is homogeneous, I felt little 
 doubt that if the principle were faithfully followed out, it 
 would lead to a solution of the problem what the nature of 
 the light which produced this e fleet was. At first I took 
 for granted that the blue light, which the prism shows 
 to be heterogeneous and not mere prismatic blue, a com- 
 pound of various colours, a little led, more green, but with a 
 
 
FLUORESCENCE. 59 
 
 predominance of blue, could only have come from light of 
 the same ref rangibility in the incident beam; but in following 
 out the consequences of that, I was led of necessity into 
 utter extravagances as regards the cause of the phenomena, 
 extravagances which did not appear to have any resemblance 
 to truth ; and on further reflection it occurred to me that 
 perhaps after all this blue colour was not produced by the 
 blue rays of the spectrum at all, but by other rays. We 
 know that the spectrum contains rays which are invisible, 
 but. in all other respects behave exactly like light. But 
 the invisibility is a mere accident, so to speak, depending 
 on the organization of the human eye : and the eyes of 
 animals in general are probably very much like the human 
 eye in this respect, although it is quite possible that certain 
 animals may see rays which we do not ; but that we cannot 
 well make out. We know that light contains besides the 
 visible rays others which are invisible ; some less refrangible 
 than the red, and others more refrangible than the violet. 
 We know that the latter show themselves especially by 
 their chemical effect, for example, on a properly prepared 
 photographic plate, and abound in sunlight and daylight, 
 which show strongly the blue light given out by quinine 
 solutions, while lamp-light, which we know to be poor in 
 those rays, shows it but feebly. 
 
 Then it occurred to me that perhaps this blue colour which 
 the solution gives out is after all the work of the invisible 
 rays which we know to accompany the visible ones. If we 
 suppose this fluid, which looks colourless like water, to be 
 excessively opaque, inky black as it were, with regard to 
 the invisible rays of high ref rangibility, and if we further 
 suppose that these invisible rays are capable of so working 
 on the fluid as to cause it to give out visible light, then 
 the explanation of epipolic dispersion and the nature of 
 epipolised light will be perfectly plain. 
 
 Now as we have been going on for some way without any 
 experiment, I will have the room darkened, and will en- 
 deavour to show you one or two. I will take the original 
 fluid in which the first phenomenon was discovered Here 
 is this green solution of leaves, and you will be able to see 
 a little of the red light. Here is the yellow glass I spoke 
 of, and in this the green band will be seen very copiously ; 
 here, again, is a solution of quinine, and with this I will 
 
60 LECTURES TO SCIENCE TEACHEK& 
 
 endeavour to show you the fundamental experiment of Sir 
 John Herschel, that this epipolic dispersion is a thing which 
 cannot be repeated ; once done the light is shorn of the 
 power of producing it. I send a beam emanating from the 
 electric lamp horizontally, and reflect it downwards on the 
 surface by a mirror. There is a second vessel of the solu- 
 tion which I place floating in the other, and the blue 
 stratum is seen at the upper surface of the fluid in the 
 upper vessel, while the quinine in the lower vessel shows 
 nothing of it. There is still the general diffused light, but 
 this intense blue stratum is wanting altogether beneath the 
 bottom of the upper vessel. Now I will replace the upper 
 beaker by one of water, when you will see the difference. 
 Just below the bottom of the upper vessel you have this 
 intense blue stratum, whereas the light passing through the 
 water in the upper beaker shows nothing of the kind. That 
 is evidence that when the blue stratum is formed once, the 
 light is shorn of the power of forming it again. 
 
 As the room is darkened I will show one or two more 
 examples of highly fluorescent solutious before I pro- 
 ceed. Here is a solution obtained in a particular manner, 
 which I will mention by and by, from the bark of the horse- 
 chestnut. 
 
 I told you what occurred to me as to the cause of the 
 phenomenon, and it was easy to test it I will not go 
 through all the experiments I tried, but pass at once to 
 what you may regard as the fundamental experiment. 
 Suppose we form a pure spectrum in the ordinary way; that 
 we reflect the sunlight horizontally into a darkened room, 
 passing it through a slit, and at a distance from the slit 
 place one or more prisms combined with a lens, so as to 
 form at the focus of the lens conjugate to the slit a pure 
 spectrum. If you were to receive the pure spectrum on a 
 screen, you would see the various colours, and if it were 
 pure enough you would see the principal Fraunhofer lines. 
 Now suppose instead of a screen you receive it on this 
 colourless fluid. The appearance is rudely represented on 
 this diagram. 
 
 I must mention that portions of the diagram are merely 
 diagrammatic, namely, the top and bottom. The middle 
 part represents what you actually see when you look down 
 on the solution, but the top represents what you would see 
 
FLUORESCENCE. 
 
 61 
 
 on a screen if you placed a screen there to receive the rays. 
 This figure represents what is seen horizontally. The rays 
 enter the solution of quinine, and the red, orange, green, 
 and blue rays pass through it as through water, and 
 come out on the other side, and if you received them 
 on a screen, you would perceive the spectrum unmodified ; 
 but when you get to about the beginning of the violet, the 
 path of the rays within the fluid is marked by a beautiful 
 sky-blue light, which at first extends right across the 
 vessel, and then the extent falls off as you get towards the 
 end of the violet. But the light does not stop there. This 
 
 FIG. l. 
 
 blue light extends far beyond the violet into a region of the 
 spectrum which contains only invisible rays ; and if the 
 spectrum is pure you see not only light there, but inter- 
 ruptions which are of the same nature as the Fraunhofer 
 lines which are seen in the visible spectrum. On a screen 
 of paper placed vertically to receive the light, the Fraun- 
 hofer lines would be seen as distinct vertical lines when the 
 paper was in focus, and would be seen a little imperfectly 
 if the paper were a little out of focus. If you imagine the 
 paper placed in one position, and moved towards or from 
 the light, any particular dark line may be considered as 
 
62 LECTURES TO SCIENCE TEACHERS. 
 
 having a locus in space, and that locus in space of one of the 
 dark lines is what you may call a dark plane. At a con- 
 siderable distance from the focus it would diverge out, 
 forming a sort of wedge. These dark planes are seen as 
 interruptions to the blue light. "What are the Fraunhofer 
 lines 1 They are parts of the spectrum when the light is 
 missing, and consequently any effect the light is capable of 
 producing will be missing too. Therefore when we get to 
 the invisible region, if the invisible light is missing the 
 visible light which that might be capable of producing will 
 be missing too, and therefore you will see interruptions in 
 this continuous mass of blue light. It constitutes a very 
 striking experiment with sunlight, when you form an 
 approximately pure spectrum by placing some prisms close 
 to a pretty broad slit, and take a tube filled with the solu- 
 tion of quinine, or a prepared solution from horse chestnut 
 bark, and make it pass through the difierent parts of the 
 spectrum in succession, beginning at the red end. At 
 first it looks like water by transmitted light ; the light 
 rays are transmitted through it as they would be through 
 water, on to the blue, but when we get on to the violet 
 then the whole of the tube is lit up with this faint 
 ghostly sort of blue light ; when you have got beyond the 
 visible rays altogether the tube is still lit up with the 
 blue light. This shows that the explanation which occurred 
 to me is really the true one, and that this blue colour is 
 produced, not by the blue rays of the spectrum at all, but 
 by other rays altogether ; that rays of one refrangibility 
 act on the fluid in such a manner as to cause it to give 
 out rays of a different refrangibility altogether, or rather, 
 of a different series of refrangibilities, because if you 
 examine a small portion of this blue light, produced by 
 rays of one definite refrangibility only, as you may do by 
 putting the solution in a pure spectrum and placing a slit 
 in front, so as to let only a narrow strip of the incident 
 rays shine on the fluid, and then analysing the narrow beam 
 from above by a prism applied to the eye, you will find 
 that the light is not homogeneous at all. Nor again is there 
 anything in the phenomenon which recalls to the mind the 
 acoustic phenomenon of harmonics ; the light is perfectly 
 heterogeneous. 
 
 I have on one of these diagrams another mode exhibited 
 
FLUORESCENCE. 
 
 63 
 
 of examining the ref rangibilit j of the light which is emitted 
 in this manner. This part is supposed to represent what 
 would be seen if you place a screen to receive the incident 
 rays, but which is not seen because you do not have a screen 
 there. Take a small portion of the spectrum, and condense 
 it further with a very small lens fixed in a blackened box, 
 so as to get a very condensed beam. This is supposed to 
 be a vessel containing one of these fluids, and the appearance 
 you get is this, but so long as you are in the visible spec- 
 trum there is, generally speaking, an image of the double 
 cone due to the light rejected from motes, which it is 
 practically impossible to get rid of ; that is followed at a 
 certain interval by a beam extending over a greater or less 
 width of the spectrum, and which is heterogeneous, that is 
 
 FIG 2. 
 
 to say contains lights of various degrees of refrangibility. 
 And if you analyse the light by a ISTicol's prism or double- 
 image prism, you will see that this speckled image is almost 
 wholly polarised in the plane of reflection, indicating that 
 it is merely due to reflected light, whereas the other, in the 
 case of a solution, is wholly unpolarised. 
 
 The phenomenon being thus explained, so far as to make 
 out the immediate nature of it, it was to be expected that 
 something of the same kind would be observed in other 
 instances of what Sir David Brewster called internal 
 dispersion, but I may mention that under that term he 
 classed together two phenomena which in reality are utterly 
 different as to their nature. In certain cases you get 
 what is virtually a powder in fine suspension, so fine that 
 
64 LECTURES TO SCIENCE TEACHERS. 
 
 it does not subside in any reasonable length of time, and 
 gives you what looks like a pretty bright solution, but 
 which really contains suspended matter. In such cases if 
 you introduce a beam of condensed sunlight, the path of the 
 beam is marked by light, because these motes reflect the 
 light which falls upon them, but that light is reflected and 
 is polarised by reflection, and this origin of the light 
 is known by its polarisation. Moreover, being simply 
 reflected it sends back the light which falls upon it un- 
 changed in kind, whereas the truly dispersed light differs 
 altogether in its nature from the light which falls upon the 
 solution, glass, or crystal that shows it. 
 
 Now as I say the phenomenon being referred to the 
 cause just explained in the one case, you may expect that 
 in other cases something similar would be perceived, and 
 I will take now the green solution obtained from leaves. 
 I will suppose the experiment exactly the same, but the 
 result is different in appearance although the nature of the 
 phenomenon is the same. The path of the rays within the 
 green fluid is marked by a blood-red light, which in 
 different parts of the spectrum penetrates to a greater 
 or less distance into the fluid. In this case the phenomenon 
 begins very near the red end of the spectrum, somewhere 
 about the line B of Fraunhofer. You first have a dart 
 of red light extending right across the vessel. Then 
 you come to a region of the spectrum for which the fluid 
 is excessively opaque, and the red light, which is produced 
 by the action of the incident light on the fluid in some 
 way or other, cannot therefore extend very far into the fluid. 
 Then you come to a region where the fluid although still 
 opaque is comparatively transparent, and the red is trace- 
 able further inwards, and so on. It is noticed, however, 
 after close observation, that where the incident rays are 
 quickly used up the red light is very copious, and where 
 they are more slowly used up the red light is not so 
 strong. 
 
 In the figure some of the red bands are slightly shaded 
 in the middle, to indicate that the red light is not so strong 
 there as elsewhere. In the case of this fluid the effect is 
 produced mainly by the visible spectrum, but it extends 
 beyond that into the invisible region beyond the violet ; 
 so that the solution of quinine, and the alcoholic solution 
 
FLUORESCENCE. 63 
 
 of the green colouring matter of leaves may, to a certain 
 extent, be regarded as extreme cases of the same general 
 phenomenon. 
 
 I will refer now to this diagram. It consists of two 
 parts ; the upper part is diagrammatic for the same reasons 
 as before ; that is, you do not see the upper half and the 
 under half simultaneously, but you may see first one and 
 then the other. When the green colouring matter is purified 
 by a particular process which would take too long to describe, 
 and you take rather a dilute solution of it, and analyse 
 the transmitted light, you get these bands of absorption. 
 When you allow the upper spectrum. to enter a very dilute 
 alcoholic solution in this manner, then the red light which 
 is* given out in the regions of the spectrum near which the 
 fluid is comparatively transparent, is given out there 
 comparatively slowly, and accordingly it is less brilliant 
 in the neighbourhood of the surface than is that corre- 
 sponding to regions of great absorption. When the fluid is 
 extremely dilute, then the intensity of the light in the 
 former part of the spectrum becomes so small that you 
 hardly see it, but its intensity where the fluid uses up the 
 incident light more quickly in producing this phenomenon 
 may still be visible, so that you may get corresponding to 
 the position of these bands of absorption in the transmitted 
 light these red bands where the light is given out very 
 copiously, the solution being so d'lute that the red light 
 given out in the intervals is hardly perceptible at all. 
 
 These phenomena are not very easily shown at a distance 
 to a large audience, but they are very striking when you 
 look at them, as two or three at a time may. in a room 
 where the sunlight is introduced or where an electric lamp 
 is used. I will, however, endeavour by and by to show 
 some of them by the aid of the electric light, but I am 
 afraid they will be seen very feebly compared to what I 
 have described, which are the effects seen when the light is 
 concentrated into a small space. 
 
 Now allow me to go back for a moment to explain one 
 particular matter I forgot at the time. I said that the 
 light transmitted through a solution of sulphate of quinine 
 was ordinary light, and that if you examined it, and 
 scrutinised especially the blue part of the spectrum, you 
 would see nothing to account for the blue colour. If you 
 
G8 LECTURES TO SCIENCE TEACHERS. 
 
 examine it by a prism applied to the eye, you will, however, 
 see that the violet, or more or less of the violet, is gone ; 
 instead of seeing as is usual the double line H, you will see 
 the fluid terminate, according to the strength of the solution 
 and the thickness looked through, more or less towards the 
 violet, say on an average about half way between the fixed 
 Knes G and H. 
 
 IsTow the incident rays work on the fluid in such a 
 manner as to cause it to give out light of a different kind 
 altogether ; a light which is found, to be heterogeneous, or 
 to consist of rays of various degrees of refrangibility. This 
 rule I find to be universal, namely, that the refrangibility 
 of the light in this process is always lowered. I have 
 never found any exception to that, nor I believe has anyone 
 since. 1 The rays which any one of these fluids is capable 
 of giving out under the influence of the^e other rays aie 
 always of lower refrangibility, and you never have the 
 refrangibility raised. 
 
 I will endeavour presently to show a test-tube with one 
 of these solutions in part of the spectrum, though I cannot 
 promise that it will be seen at a distance. The fact is I 
 am accustomed to work with sunlight rather than with the 
 electric light and I require more preliminary trials than I 
 allowed myself for making the thing succeed. Still I think 
 you see that on interposing a test-tube with the solution of 
 quinine in the beam from the electric lamp, after it has 
 passed through the prism, it cuts off certain portions of the 
 spectrum thrown on the wall beyond, forming a shadow 
 which shows in what part of the rays proceeding to form 
 the spectrum the tube is for the moment placed ; the blue 
 light with which the solution glows, commencing about the 
 violet, is seen altogether beyond the region of the visible 
 rays. Here is a solution of a substance obtained from the 
 bark of the horse-chestnut which shows it still better. You 
 observe the blue band beyond the visible spectrum altogether. 
 Another instance is when we allow the beam of light to fall 
 on a piece of red cloth, it shows an orange band beyond the 
 visible rays. 
 
 1 Calorescence, or tho exhibition of light by a body intensely heated 
 by .the concentration upon it of invisible heat-rays, is in some respects 
 so different from the phenomena of fluorescence or phosphorescence that 
 I do not regard it as forming any exception to the rule. 
 
FLUORESCENCE. 67 
 
 I have been a little anticipating what was to come, 
 namely, that these phenomena are not confined to fluids or 
 clear solids, but that they can be seen in every case. I 
 have shown you that in a spectrum if you separate out the 
 rays from one another by prismatic refraction you can see 
 the phenomena in the invisible rays. But there is another 
 mode of separating light into two portions of which one 
 is allowed to pass, which is easier in practice, and which 
 exhibits some of these phenomena very beautifully that 
 is by analysing it by absorption. For instance, I have here 
 a very deep blue glass which cuts out most of the visible 
 light, but it admits the violet and certain invisible rays 
 beyond very copiously. It is a cobalt blue glass, and here 
 is a yellow glass. I will analyse the light, not by a prism, 
 but simply by absorption. This jar at present contains 
 nothing but water. Now I will put the blue glass on to 
 the electric lamp, and I have here a solut: on obtained from 
 the bark of the horse chestnut, a little of which I will drop 
 into the jar. If you make a decoction of the bark, which 
 contains a good deal of tannin, the solution soon becomes 
 brown. It contains however two crystal lizable substances, 
 called esculin and fraxin, which can be obtained chemically 
 pure. The alkaline solutions of these bodies have this 
 power in a high degree. You can get rid of the tannin by 
 making a decoction, and when cold adding some persalt of 
 iron or salt of alumina, precipitating by ammonia, and 
 filtering, and you then get this beautiful solution, which 
 will keep very fairly. I will pour some of this into the 
 water, and as it sinks down it forms a beautiful blue cloud. 
 If I hold the blue glass so as to intercept the incident rays 
 hardly any diminution will be perceived, while the yellow 
 glass cuts off most of the rays by which the effect is pro- 
 duced ; but if I put the yellow glass between the jar and your 
 eyes, a great deal of the light is transmitted The general 
 effect is shown very well by means of these glasses. Here 
 also are some jars containing some fluorescin and other fluids 
 which can be tried in the same way. A coloured medium will 
 absorb in a different manner the rays that fall on the fluid arid 
 the rays coming from it. That leads to one method of obser- 
 vation, which does not require the apparatus which I have 
 hitherto supposed, but is exceedingly simple and at the same 
 time very effective. You do not even require sunlight, 
 
 F 2 
 
63 LECTURES TO SCIENCE TEACHERS. 
 
 you can work with ordinary daylight. Suppose you have 
 a room which you can darken, and that you are at liberty 
 to cut a hole four or five inches square in the shutter, under 
 which it is convenient to screw on a ledge for the sake of 
 supporting the object to be examined, and that you cover the 
 hole by a suitable glass. The most useful generally is a 
 dark blue coloured by cobalt, or a dark violet coloured by 
 manganese. The blue does better for some things and the 
 violet for others. Cover the hole in the shutter with the 
 deep coloured glass or, as is occasionally better, with a 
 solution of a salt of copper, such as the nitrate, which is 
 more convenient than the sulphate on account of its great 
 solubility. Suppose you have daylight filtered as it were 
 through the deep blue or violet glass ; then if you place 
 in front of the glass a test tube containing a solution of 
 quinine in dilute sulphuric add, or a solution obtained from 
 the bark of the horse-chestnut with a little ammonia, you 
 will see this blue phosphorescent looking light to perfection. 
 But supposing you have not a window-shutter which you 
 can make a hole in, still you can get on very well with an 
 old packing-case. You knock off part of the top of it, so 
 that you may look in, and saw off a portion obliquely 
 parallel to the opposite top edge and on the slanting rectan- 
 gular hole thus formed you nail a piece of board, making 
 a window in it four or five inches square, with a little 
 ledge to keep the glass with which you cover it from 
 slipping down You place it near the window and cover 
 your head with a dark cloth as if you were looking into a 
 camera obscura, and so you can see the phenomenon to per- 
 fection. 
 
 If you want to demonstrate that it is really this phenomenon 
 you are dealing with, it is desirable to have a second glass 
 in a certain sense complementary to the first. If we could 
 pick out media which absorbed light just in the way we 
 wished, we should choose a coloured glass perfectly opaque 
 from the red end up to the blue or violet, and perfectly trans- 
 parent beyond, and a second glass perfectly opaque for those 
 rays for which the first was transparent, and vice versa. 
 But as we cannot make media to command to absorb what 
 parts of the spectrum we like, we must make use of the 
 best which the colouring matters of nature afford us ; and 
 if you take a blue glass and a yellow glass they will in 
 
FLUORESCENCE. 69 
 
 most cases answer the purpose sufficiently well. Suppose 
 then you have a dark glass on a window-shutter and you 
 have in front of it a substance to be examined, and it gives 
 out this beautiful phosphorescent light. In general this 
 may be at once distinguished* from mere scattered light ; 
 but to make sure of it use your yellow glass, and place 
 it between the blue and the substance you are examining. 
 If it is well chosen it will cut off almost all the effect. 
 Then place it between your eye and tjie medium which 
 is shining with this phosphorescent light and you will 
 see it quite plainly. The difference of effect with this 
 additional glass in the two positions proves that you 
 have really to deal with this peculiar phenomenon, and 
 enables you to at once distinguish it from some appearances 
 which at the first glance resemble it very much. If you 
 put a minute quantity of a solution of proto- chloride of tin 
 into a large quantity of common water, the mixture will 
 have a bluish look by reflected light, and if you condense 
 sunlight upon it you will get a beam somewhat like what 
 you do when you receive a beam on the solution of sulphate 
 of quinine. This however is merely scattered light ; and 
 that it is so is shown at once when you come to make 
 experiments upon it in which you strain the incident light. 
 For example, if you place the mixture inside your darkened 
 chamber in lieu of the solution of quinine, the difference 
 will appear in a moment. The mixture will merely give 
 out a little light of a deep blue colour, which is scattered 
 light, whereas the solution of quinine will be lighted up 
 with this beautiful light that you see. I may mention a 
 very simple and pretty experiment which can be made in 
 that way. Take a bit of common horse-chestnut bark, 
 float it in a glass of still water in which a drop of ammonia 
 had been mixed ; the peculiar substances contained in 
 the bark will begin to be dissolved, the solution will descend, 
 and you will see streams of descending blue light. If you 
 can obtain specimens of these substances esculin and 
 fraxin, a minute quantity of the two thrown together on 
 the water instead of the bark looks very pretty. The 
 substances will form little luminous specks here and there 
 on the surface, which will give rise to descending streams 
 of blue and greenish light. 
 
 Now there is another way of testing the change of 
 
70 LECTURES TO SCIENCE TEACHERS. 
 
 refrangibility almost without apparatus, by using your 
 darkened chamber with a piece of blue glass in the window. 
 Suppose that in front of the blue glass you place a piece of 
 white earthenware, such as a saucer turned upside down. 
 If you hold a slit at arm's length and view it through a 
 prism, in the first instance aiming at the blue glass and 
 looking up at the sky, you will see the sort of light trans- 
 mitted. The brighter parts of the original spectrum will be 
 almost entirely wanting, but you will see the violet, much 
 of the blue, and the faint extreme red which is freely trans- 
 mitted, and if the glass be not very deeply coloured a little 
 faint greenish yellow which is not yet wholly absorbed. 
 If you now aim at the white plate instead of the sky, you 
 will see just the same spectrum as before, only not quite so 
 strong. Now suppose you lay on the white plate a little 
 bit of ordinary scarlet cloth, hold the slit close to that, and 
 aim at both the cloth and the white plate, so as to get 
 from different parts of the slit a spectrum of the light 
 coming from each. This particular cloth in the blue field 
 would look red On examining the joint spectrum, the part 
 seen by projection of the slit on the plate will appear as 
 just described, while that seen by projection on the scarlet 
 cloth will show a prolongation of the extreme red, and a great 
 deal of bright light where there is none in the incident light, 
 while the violet part will be nearly black. 
 
 This diagram, which was made for another purpose, may 
 serve to exhibit what you would see in that particular case. 
 It really represents what you see by looking through a 
 crystal of nitrate of uranium placed immediately behind 
 the slit with which you aim at the white light sifted 
 through a blue medium. There are certain bands of 
 absorption where there is a maximum of opacity in the 
 incident light ; and when you analyse the beam of light 
 which comes through you see in the transmitted rays there 
 are certain dark bands of absorption ; but over and above 
 that, there is light created with a refrangibility less than 
 exists at all in the incident beam, and in the particular case 
 of the salt of uranium the prismatic composition of this light 
 is very peculiar, its spectrum consisting of bright bands. 
 
 Now as I want to show you how to make experiments 
 yourselves without apparatus, I may mention, that suppos- 
 ing you have one of these solutions in a test tube, you very 
 
FL UOEESCENCE. 7 1 
 
 much increase the effect by plunging the test tube in water 
 and looking at it downwards nearly parallel to the test- 
 tube. The reason of that is that the incident rays fall 
 upon the fluid and cause it to give out light in all direc- 
 tions, blue light, or whatever else it may be, according to 
 the nature of the fluid ; but when the water is not there, a 
 portion "of the light so given out does not enter the eye at 
 all, but suffers total internal reflection at the outer surface 
 of the tube. On the other hand, if you look inside the tube 
 the light suffers absorption on the part of the fluid itself. 
 It does not much signify in the case of sulphate of quinine, 
 which is sensibly transparent, but if you examine any 
 coloured fluorescent fluid looking down the tube inside, 
 you lose light from defect of transparency, the light being 
 absorbed by the fluid ; whilst if you look outside you lose a 
 great deal by the total internal reflection at the surface of 
 the glass. But if you plunge the test-tube into water, and 
 look down from the outside, any of the emitted light which 
 gets into the glass of the tube is able to get out again, so 
 that you can look down from above in a very slanting 
 direction and still get all the light, and as the stratum 
 which emits the light is seen very much foreshortened, the 
 brightness of the light is thereby increased. 
 
 Now as to the cause of this phenomenon. From the first 
 I believed the cause to be this : that the incident rays so 
 act on the ultimate molecules of the body as to throw them 
 into a state of agitation, which agitation they in their turn 
 are capable of communicating to the ether. Everybody 
 now, I believe, considers that light is produced by the 
 vibration of a certain subtle medium, which we call the 
 luniiniferous ether, and I will take a dynamical illustration 
 of the phenomenon according to this view. Suppose you 
 had a number of ships at rest on an ocean perfectly calm. 
 Supposing now a series of waves, without any wind, were 
 propagated from a storm at a distance along the ocean, 
 they would agitate the ships, which would move backwards 
 and forwards ; but the time of swing of the ship would 
 depend on the time of its natural oscillation, and would 
 not necessarily synchronise with the periodic time of the 
 waves which agitated the ship in the first instance. The 
 ship, being thus thrown into a state of agitation, would itself 
 become a centre of agitation, and would produce waves 
 
72 LECTURES TO SCIENCE TEACHERS. 
 
 which would be -propagated from it in all directions. This 
 I conceive to be a rough dynamical illustration of what 
 takes place in this actual phenomenon, namely, that the 
 incidence of etherial waves causes a certain agitation in the 
 ultimate molecules of the body, and causes them to be in 
 their turn centres of agitation to the ether ; in fact that 
 the incident light renders the medium so to speak self- 
 luminous, so long as it is under the excitement of the 
 incident light. That is the view which I maintained from 
 the first, and which is clearly expressed in my original 
 memoir, which was published in the Philosophical Trans- 
 actions of 1852. There is one phenomenon, that of phos- 
 phorescence, which I felt from the first to be exceedingly 
 analogous to that which is now known by the name of 
 fluorescence, a word I suggested in that original memoir, 
 derived from fluor spar, which was one of the first minerals 
 in which the phenomenon had been observed, as the 
 analogous term, opalescence, is derived from the name of 
 the mineral opal. I am unable to draw any sharp line of 
 demarcation between fluorescence and phosphorescence. So 
 far as I had observed, the effect was only of instantaneous 
 duration, although, as I have expressly stated, I had not 
 made experiments on a revolving mirror to determine 
 whether a finite duration could be perceived With regard 
 to the explanation of the law which I believed to be 
 universal, that in this phenomenon the refrangibility is al- 
 ways lowered, that is to say, the light coming out is always 
 of lower refrangibility than the incident light, I offered 
 a certain conjecture, which I did not hold to very tightly, 
 and I have somewhat changed my views in that respect ; 
 but I held from the first that the effect is not a direct . 
 but an indirect one ; that the light is not simply reflected 
 from the ultimate particles of bodies. It is curious that 
 some two or three writers have attributed to me the notion 
 that in this phenomenon the light reflected from the mole- 
 cules of the body was changed in refrangibility. They have 
 attributed that notion to me, and then contended against it ; 
 but if you will allow me to read a short passage from my 
 original paper, it will show that I am not responsible for 
 that. I wrote these words : " In considering the cause of 
 internal dispersion, we may, I think, at once discard all 
 supposition of reflections and refractions of the vibrations 
 
 
FLUORESCENCE. 73 
 
 of the luminiferous ether amongst the ultimate molecules 
 of bodies. It seems to me quite contrary to dynamical 
 principles to suppose that any such causes should be 
 adequate to account for the production of vibrations of one 
 period from vibrations of another." Having written that, 
 I am not responsible for the view which has been so 
 wrongly attributed to me. I can only account for it in 
 this way. I suppose it was from the title I gave my paper, 
 " On the change of refrangibility of light," which I chose 
 because undoubtedly the most striking part of the 
 phenomenon, which had not been hitherto suspected, was 
 that the light given out was of a different refrangibility 
 from the light going in. With regard to the duration of 
 the phenomenon, I thought it possible that, though very 
 large compared with the time of a single luminous vibration, 
 it might elude our means of observation ; but subsequently 
 Mons. E. Becquerel showed, by a very ingenious instrument, 
 which is here on the table, which he calls a phosphoroscope, 
 that the phenomenon is really of appreciable duration. 
 The light enters at one side, and falls on -a cell containing 
 the substance to be examined, in this case a salt of 
 uranium enclosed between a pair of discs constituting a 
 double fly with holes of a somewhat sectorial shape, so 
 arranged that the holes in the one exactly correspond with 
 the spaces between the holes in the other. When you turn 
 the wheel round, supposing there is no substance there, 
 you see no light, because you are always looking across a 
 plate of metal. At one time there is a hole in the front 
 disc, which is covered by the second plate, and when there 
 is a hole in the second plate it is covered by the front disc ; 
 a series of holes coming alternately in the two discs. But 
 supposing a substance is interposed, and the light is let in, 
 and you turn the wheel, the illumination is let on and cut 
 off alternately with great rapidity ; but you see the light, 
 not at the moment when the body is illuminated, but a 
 small fraction of a second afterwards, through the hole 
 which is opposite your eye ; in that way you see it a very 
 short space of time after it has been lit, so to speak, by the 
 incident light ; it gets a number of doses of light in one 
 rotation, and you get a number of glimpses of it immediately 
 afterwards. In that way many substances which show this 
 phenomenon appear luminous, and you see them by the light 
 
74 LECTURES TO SCIENCE TEACHERS. 
 
 which has been treated in this manner. I may mention 
 also that even a simple revolving mirror will show the 
 duration of the effect in such cases as the salts of uranium 
 and solids in general. If you use an ordinary electric 
 machine as a source of illumination, giving a succession of 
 sparks, or an induction coil with a Leyden jar in connection 
 with the two terminals, which gives a momentary discharge, 
 you can observe the substance in a rapidly revolving mirror, 
 and in that way you get a momentary view of the substance 
 by reflected light, while the illumination due to fluor- 
 escence, in case it has an appreciable duration, is drawn out 
 into a broad gleam ; so that even without an instrument of 
 the kind now on the table the duration of the effect can 
 be manifested by experiment. The duration of the elfect, 
 I may observe, has not as yet in any instance been 
 demonstrated experimentally in the case of a liquid. 
 
 When epipolic dispersion wa,s referred to a change of 
 refrangibility of light, there were some older experiments 
 which at once received their explanation. For instance, 
 Sir John Herschel himself, on throwing a pure spectrum 
 on turmeric paper, had noticed a great prolongation of the 
 ordinary visible spectrum. But he supposed that this was 
 due to the ultra violet rays, which were directly reflected 
 into the eye ; and he speculated whether there might not be 
 a repetition of the colours of the ordinary spectrum. This, 
 however, proved to be a phenomenon of the kind I have 
 just described, fluorescence, or, if you like so to call it, 
 phosphorescence, as I am persuaded that fluorescence is 
 nothing but phosphorescence of brief duration. Mons. E. 
 Becquerel went still nearer to the actual phenomenon, 
 for he was making experiments on substances which show 
 phosphorescence after exposure to light, and observed that 
 some of them were specially luminous when light fell upon 
 them and was acting upon them. Nevertheless, although he 
 correctly explained what he witnessed in these cases, from 
 connecting it too closely with phosphorescence, he failed to 
 perceive the full bearing of his own observation ; and 
 though he had actually under his hands the solution of 
 quinine, and had discovered by means of photography 
 the intense absorbing action of that fluid on the invisible 
 rays, and expressly mentioned the " dichroism " of the solu- 
 tion he never dreamt of putting the two things together, 
 
FLUORESCENCE. 75 
 
 and showing that the peculiar coloured light exhibited 
 in this and other allied instances, which were matters of 
 ordinary observation, 1 had an origin hitherto unsuspected 
 
 I have here some phosphorescent tubes which have 
 been lent to me, which you will see retain the luminosity 
 for some time in the dark after the light of the electric 
 lamp has been allowed to play upon them. The most 
 phosphorescent substances are certain sulphides of metal-s 
 of alkaline earths, though it is said that certain impurities 
 contribute to the effect rather than otherwise. 
 
 In the course of my experiments I was led to see that 
 glass was by no means transparent in regard to the most 
 refrangible of the rays I had to deal with. I procured 
 accordingly prisms and lenses of quartz, with which to form 
 a pure spectrum. On applying them to the solar spectrum, 
 the invisible rays were seen to extend far beyond anything 
 I had ever seen before, and showed a continuation of 
 Fraunhofer's lines that is represented on these maps, 
 which were originally drawn for an evening lecture I 
 gave on the subject for the British Association at 
 Belfast in 1852. In the spring of the next year, while 
 preparing for a lecture at the Royal Institution, in the 
 laboratory of that institution, along with Faraday, though 
 I had expected beforehand to obtain a very long spectrum 
 by the use of the electric light, I was utterly surprised 
 when I found the actual length of it. I used in the first 
 instance a Ley den jar, and I could not but think at the first 
 moment that there was some stray reflection of light, for if 
 the visible spectrum was about one inch, the whole spectrum 
 obtained by means of powerfully fluorescent substances, 
 such as uranium glass, with the electric light, was about 
 six or eight inches, which was a length I had not been at all 
 prepared for. 
 
 1 The very name " Schillerstoff, " formerly given to sesculin, is 
 derived from the property in question. 
 
THE KINEMATICS OF MACHINERY. 
 
 ILLUSTRATED BY THE BERLIN COLLECTION OF KINE- 
 MATIC MODELS. 
 
 TWO LECTURES. 
 
 BY PROF. ALEX. B. W. KENNEDY, C.E., OF UNIVERSITY COLLEGE, 
 LONDON. 
 
 T/^e substance of the following pamphlet formed two lectures 
 delivered to Science Teachers at South Kensington during the month of 
 August 1876. They have been recast only so much as to make them 
 intelligible without the aid of the set of models by which they were then 
 illustrated. Any readers' who may wish further to study Rculcaux's 
 treatment of the subject, of which I have attempted here to no'e a few 
 salient points, I must refer to his own " Theoretische Kinema'ik," or to 
 my translation of it, published under the title of " The Kinematics of 
 Machinery. " In th'is work the whole matter is taken up in great detail 
 from the point of view which I have endeavoured to indicate. 
 
 A. B. W. K. 
 LECTURE I. 
 
 MOST of the models used to illustrate this and the follow- 
 ing lecture belong to the Kinematic Collection of the Gewerbe- 
 Akademie in Berlin, and have been designed by Professor 
 Keuleaux, who is the Director of the Academy and a 
 Professor in it. The rest were sent to the Loan Collection 
 by Messrs. Hoff and Voigt of Berlin, and Messrs. Bock and 
 Handrick of Dresden. In essentials there is no difference 
 between the Berlin and the Dresden models. Both have 
 been designed specially for use in instruction in the Kine- 
 matics of machinery. 
 
 I must first try to explain briefly but exactly what I 
 mean by the phrase " Kinematics of machinery." Professor 
 Keuleaux, whose models are before us, defines a machine as 
 " a combination of resistant bodies so arranged that by their 
 means the mechanical forces of nature can be compelled to do 
 work accompanied by certain determinate motions." The 
 complete course of machine instruction followed in some of 
 the continental technical schools covers something like 
 the following ground : 
 
 First, there is the perfectly general study of machinery, 
 
THE KINEMATICS OF MACHINERY. 77 
 
 technologically and teleologically. Then there comes what 
 we may call the study of prime-movers, which in terms of 
 our definition would "be the study of the arrangements by 
 means of which the natural forces can be best compelled to do 
 the required work. Then comes the study of what may be 
 called " direct-actors," or the direct-acting parts of machinery ; 
 in the terms of our definition, the arrangement of the parts 
 of a machine in such a way as best to obtain the required result. 
 Next comes what we call machine design ; the giving to the 
 bodies forming the machine the requisite quality of resist- 
 ance. Machine design is based principally on a study of 
 the strength of materials. 
 
 One clause of the definition still remains untouched. The 
 machine, we said, does work accompanied by certain deter- 
 minate motions. Corresponding to this we have in machine 
 instruction the study of those arrangements in the machine by 
 which the mutual motions of its parts, considered as changes of 
 position only, are determined. The limitation here must be 
 remembered ; motion is considered only as change of posi- 
 tion, not taking into account either force or velocity. This 
 is what Professor Willis long ago called the " science of pure 
 mechani-m," what Eankine has called the " geometry of 
 machinery," what Reuleaux calls '''kinematics," and what 
 I mean now by the " kinematics of machinery." 
 
 The resets of many years' work of Eeuleaux in connection 
 with this subject are embodied in his book Die Theoretische 
 Kinematik, which I recently had the pleasure of trans- 
 lating, and I shall endeavour to give you an outline* of his 
 treatment of the subject. It cannot be more than an outline, 
 as you will readily understand. The subject is a very large 
 one, and I have had to choose between taking up many branches 
 of it and merely mentioning each, and confining myself to a few 
 points, and going more into detail about them. I have chosen 
 the latter plan, believing that the former would be of little 
 benefit to anybody. It will be easy for those who are 
 sufficiently interested in the matter tc follow it up, and to 
 study those parts which I omit by the aid of the book I 
 have just mentioned. My lecture to-day will be principally 
 theoretical, and to-morrow I shall go more into practical 
 applications. So far as possible, as I have Professor 
 Keuleaux's models before me, I shall endeavour to follow his 
 own order in treating the subject. 
 
78 LECTURES TO SCIENCE TEACHERS. 
 
 I presume you are acquainted to a certain extent with the 
 ordinary method of studying "pure mechanism;" the method 
 originated by Monge (1806), developed in Willis' well-known 
 Principles of Mechanism (1841) and made popular to a 
 great extent by Prof. Goodeve's capital little text-book 
 and others. Each mechanism is studied for and by itself, 
 in general by the aid of simple algebraic or trigonometric 
 methods, and is spoken of in reference to a certain " con- 
 version" of motion which occurs in it. Thus we have the 
 conversion of circular into reciprocating motion, the con- 
 version of reciprocating into circular, &c., and simple formulae 
 express certain relations between the motions of two or more 
 moving points. In this way we know something important 
 about a great number of mechanisms, and arrive at many 
 results which are both useful and interesting. Some things 
 are still left wanting, however ; and these things may be 
 summed up in this way : 
 
 (1) We notice at once that we have taken the mechanism 
 as a whole. We do not analyse it in any way whatever, 
 and therefore, 
 
 (2) We have scarcely any knowledge of its relations with 
 other mechanisms, or (what is quite as important) of the 
 various forms which one and the same mechanism may take. 
 We shall see presently how extraordinarily various these forms 
 are. We have never a general case with special cases 
 derived from it ; each case is treated by itself as a special 
 one. Then 
 
 (3) 'The mechanism is studied in general from a point of 
 view which gives us only the conditions of the motion 'of 
 two points in it, or two portions of it, and is then left. 
 The kinematic conditions of the mechanism as a whole 
 remain absolutely untouched. 
 
 In such a mechanism as that of an ordinary steam-engine, 
 for instance, we study the relative motions of the guide- 
 block and the crank, or I ought perhaps to say of the axes 
 of the crosshead and of the crank-pin. We thus know the 
 motions of two points in the rod which connects those axes, 
 the "connecting-rod," but we leave the motions of all its 
 other points untouched. It may, of course, be said that 
 these others are of much less practical importance. This is 
 true to some extent, although their practical importance is 
 greater than might be supposed at first. But in any case 
 
THE KINEMATICS OF MACHINERY. 79 
 
 these motions must certainly be studied if we are to obtain 
 a complete knowledge of the mechanism to which they 
 belong. Any method of study, therefore, which covers all 
 the kinematic conditions of the mechanism, instead of the 
 mechanical conditions of two or three points only, possesses 
 in that respect very great advantages. 
 
 The treatment of mechanisms which I shall sketch to you 
 is intended to remedy some of the defects which I have 
 enumerated. Those of you who have studied modern 
 geometry side by side with the old methods will recognise 
 that these defects are somewhat analogous to those of 
 Euclidean geometry. The attempt to remedy them proceeds 
 in lines similar to those of modern geometry, and will 
 eventually, I believe, when more fully worked out, take 
 the same position in i's own subject. 
 
 Let us then look first at the analysis of mechanisms. This 
 is none the less important a matter that its results are so 
 very simple in many cases. A clear understanding of these 
 elementary matters is of great assistance in clearing up 
 difficulties which occur in the more advanced parts of the 
 subject. 
 
 In a machine or a mechanism of any kind the motion of 
 every piece must be absolutely determinate at every instant. 
 Jt will be remembered that we are at present considering 
 motion as change of position only, not in reference to velocity. 
 The motion or change of position may be determined by the 
 direction arid magnitude of all the external forces which act 
 on the body : the motion is then said to be free, but it is 
 obviously impossible to arrange such a condition of things in 
 a machine. The motions may, however, be made absolutely 
 determinate independently of the direction and magnitude 
 of external forces, and in order that this may be the case 
 the moving bodies, or the moving and fixed bodies as the 
 case may be, must be connected by suitable geometric forms. 
 Motion under these circumstances is called constrained motion. 1 
 
 If I allow a prismatic block to slide down the surface of 
 an inclined plane its motion will be free ; it is determined 
 by the combination of external forces which act upon the 
 
 1 Essentially it dees not differ from free motion ; the difference 
 really lies in the substitution of stresses or molecular forces, which are 
 under our complete control, for external forces. 
 
80 LECTURES TO SCIENCE TEACHERS. 
 
 block. If the block be pressed on one side as it slides, it at 
 once moves sideways, and can only be kept in a straight path 
 if directly the pressure is exerted on the one side an equal 
 and opposite force (or a force which has a resultant with the 
 first in the direction of motion), be caused to act upon it on 
 the other. If, on the other hand, the block be made to slide 
 between accurately fitting grooves (like a guide-block in a 
 machine), inclined at the same angle as the plane, and like 
 it fixed, the block may be pressed sideways or in any other 
 direction, but no alteration in its motion can take place ; the 
 motion is "constrained," it can occur only in the one 
 direction permitted by the guiding grooves. In the one 
 case the external force has to be balanced by another 
 external force, in the other the balancing force is molecular, 
 i.e. is a stress and not an external force, and comes at once 
 into play the instant the disturbing force is exerted. The 
 geometric forms which are used in this way to constrain or 
 render determinate the motions in machines are very various, 
 and are chosen in reference to the particular motion required. 
 If every point in a body be required to move in a circle 
 about some fixed axis, a portion of the body is made in the 
 form of a solid of revolution about that axis, and this is 
 caused to " work in " another similar solid ; the two form- 
 ing the familiar pin and eye. If all points of a body be 
 required to move in parallel straight lines we get similarly 
 for guiding forms a pair of prisms of arbitrary cross- section ; 
 a slot and block. If every point of a body be required to 
 move in a helix of the same pitch we use a pair of screws 
 of that pitch, one solid and one open, for constraining the 
 motion ; a screw and nut. 
 
 The general condition common to these very simple forms 
 is that in each case the path of every point in the moving body 
 is absolutely determined at every instant, that is to say, the 
 change of position of the moving body is absolutely 
 determinate. 
 
 The geometric name for these mutually constraining bodies 
 is envelopes, and each one is said to envelope the other. We 
 shall call them (kinematic) elements, and the combination of 
 two of them we shall call a pair of elements. 
 
 Those we have mentioned are special and very familiar and 
 important cases of pairs of elements, which are of great 
 simplicity. They have the common property of surface 
 
THE KINEMATICS OF MACHINERY. 
 
 81 
 
 contact, the one enclosing the other, and are therefore called 
 closed or lower pairs of elements. They are, moreover, the 
 only closed pairs which exist. They are further the only 
 pairs in which all points of the moving element have similar 
 paths. 
 
 Every point of an eye, for instance, moves in a circle about 
 the same axis. If there were attached to it a body of any 
 size or form whatever, all its point would move about the 
 same axis. The " point-paths " would all be concentric circles. 
 Again, whatever the external size or shape of a nut, every 
 point in it moves in a helix of the same pitch about the axis 
 of the screw ; the point-paths, that is, would be similar. 
 
 The general condition of determinateness of motion can, 
 however, be fulfilled by an immense number of other pairs of 
 elements. The theory of these is too large a subject to be 
 entered into just now, I must merely direct your attention to 
 the existence of such combinations. 
 
 u 
 
 Fio. 1. 
 
 Fig. 1 represents one of the simplest that can be used. 
 Here one of the elements is an equilateral triangle, ABC, 
 the other is the "duangle" R P S Q. The latter moves 
 within the former, touching it always in three points, or rather 
 along three lines. Its motion is just as absolutely determinate 
 as the motion of a pin in an eye. It is free to move at 
 any instant only about the point in which the three normals 
 to the triangle at the points of contact intersect (as ,Q in the 
 Fig.) The models before you show a few of the many forms 
 
82 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 taken by such pairs of elements. It is worth while noticing 
 a few points in which the motions determined by them differs 
 from the motions of the closed pairs. First, as we have 
 already seen, the contact of the elements determining the 
 motion was surface-contact in the former case, while here it 
 takes place only along a finite number of lines. Then the 
 motions of all points in the first case were similar ; in these 
 pairs the motions of the points are not similar, but entirely 
 
 Flo. 2. 
 
 dissimilar, the motion of each point depending entirely upon 
 its position. Fig. 2 shows a few of the point-paths of the pair 
 of elements shown in Fig. 1. The strikingly different curves 
 obtained from one pair of elements, according to the choice 
 of the describing point, is too obvious to need further notice. 1 
 
 1 The triangle UTQ and the three curves within it, which have M^ 
 for their centre, are point-paths. The curve -triangle and the duangle 
 shown in thicker lines will be explained further on. 
 
THE KINEMATICS OF MACHINERY. S3 
 
 These pair of elements are called higher pairs. They have 
 only a few applications in practice, their interest being chiefly 
 theoretical. From our present point of view their theoretic 
 interest is considerable, because of their exact analogy with 
 the lower pairs. 
 
 There is another difference between the two kinds of pairs 
 which deserves notice, for reasons which will be better under- 
 stood afterwards. The pair of elements determines the rela- 
 tive motion of the two bodies connected by it. If one body 
 be stationary on the floor or the earth, the moving body has the 
 same motion relatively to the floor or earth that it has to the 
 other element. If I move about both bodies in my hand, both 
 have motion relatively to the earth, but the relative motion 
 of the one to the other remains unchanged. It is of course only 
 a case differing in degree from the former one, for in the 
 former one both bodies had the motion of the earth itself, 
 while one had the additional motion which I gave it. We may, 
 however, not to be pedantic, speak of anything as "fixed," or 
 ** stationary " which has the same motion as the earth. 
 
 Now (in this sense) we may fix either element of a pair, and 
 with the lower pairs the relative motion taking place remains 
 the same whichever element be fixed. With the higher pairs, 
 on the other hand, the relative motion is altered, and the point- 
 paths become entirely different. The point-paths of the 
 duangle relatively to the triangle are, for instance, quite dif- 
 ferent from those of the triangle relatively to the duangle. This 
 change of the fixed element is called the inversion of a pair. 
 The ultimate result of our analysis of mechanisms is then 
 pairs of elements ; we cannot go below this. The pairs we 
 have noticed are of two kinds, each having their own 
 definite characteristics. If now two or more elements of as 
 many different pairs be joined together we get a combination 
 which is called a (kinematic) link. It is obvious that the 
 form of such a link is, kinematically, absolutely indifferent. 
 The choice of its form and material belongs to machine- 
 design. It may be brick and mortar, cast-iron, timber, as 
 we shall see afterwards, but the fact that this is indifferent 
 kinematically cannot be too distinctly kept in mind. 
 
 We can make combinations of links by pairing the elements 
 which each contain to partner elements in other links, and 
 such combinations are called kinematic chains. Thus if we 
 denote similar elements by similar letters, aa, bb cc, &c. and 
 
 o 2 
 
84 LECTURES TO SCIENCE TEACHERS. 
 
 the link connection by a line, we may indicate some of the 
 chains obtainable from 4 pairs and 4 links, thus : 
 
 a bb cc dd a 
 
 (we suppose the " chain " to return on itself and the two ele- 
 ments a to be paired, the whole forming a closed chain) ; or, 
 
 a cc bb dd a 
 
 or 
 
 a dd cc bb a &c. 
 
 FIG. 3. 
 
 For the sake of illustration we give in Fig. 3 a sketch 
 of a familiar chain containing four links, each connected to 
 the adjacent link by a cylinder-pair of elements. The axes 
 of the four pairs of elements are parallel. 
 
 We have then, in the kinematic chain a combination so 
 constructed that all its parts have determinate motions, 
 motions absolutely fixed by the form of the elements carried 
 by its links, and independent (considered as changes of posi- 
 tion) of the application of external force. To convert the 
 chain into a mechanism we have only to do what we have 
 already done in connection with pairs of elements, fix one 
 element, or, as each element is rigidly connected with a link, 
 we may say preferably fix one link. Any link may be fixed, 
 the chain therefore gives us as many mechanisms as it has 
 links. In general these are different, in special cases only 
 two or more of them are the same. We shall be able to enter 
 into this part of our subject at some length in the next lecture ; 
 at present it will suffice to note two or three of the leading 
 characteristics of chains and mechanisms which we can now 
 easily recognise. These are 
 
 (i.) That the motion of any link relative to either adjacent 
 link is determined by the pair of elements connecting them. 
 
 
THE KINEMATICS OF MACHINERY. 85 
 
 (ii.) That the motion of any link relative to any other 
 than its adjacent links depends on all the elements of the 
 chain. 
 
 (iii.) That no one link of a mechanism can be moved with- 
 out moving all the other links except the fixed one, and 
 
 (iv.) That there can be only one fixed link in a mech- 
 anism. 
 
 The two last propositions require a few words of explana- 
 tion. Suppose that in any combination of, say, four links, 
 two can be moved without moving the other two, the com- 
 bination is actually one of three links only, for clearly the 
 two immovable links may be made into one, and are two 
 only in name. This is very often the case in machinery, 
 where special mechanisms are frequently used for the express 
 purpose of connecting rigidly two or more links, and making 
 them act as one, at certain intervals. 
 
 If, however, in the combination supposed, one link be fixed, 
 while two can be moved and the fourth can either move or be 
 stationary, the combination no longer comes under our defini- 
 tion of constrainment, for the motions are at a certain point 
 indeterminate, at the point, namely, when it is possible for 
 the fourth link either to move or to stand. Chains often 
 occur in which this would be the case were it not that mecha- 
 nicians take means, either by adding other chains or in other 
 ways, to constrain the motion which would otherwise be 
 useless to them. 
 
 We have now obtained some idea of the way in which 
 mechanisms are formed, of the elements of which they consist. 
 Before applying the knowledge we have thus acquired I must 
 direct your attention to some geometric propositions which 
 will greatly facilitate the theoretic dealing with these 
 mechanisms. 
 
 In order that I may not enter into too wide a subject, I 
 shall confine myself here to the consideration only of " con- 
 plane " motions, or motions in which all points of the moving 
 body move in the same plane or in parallel planes. The 
 limitation is a large one, but the cases included under 
 conplane motion cover the greater part of those which occur 
 in practice. The method I have to describe is equally 
 applicable to general motion in space as to simple constrained 
 conplane motions of which I shall speak. 
 
 Let me remind you that the motion of any figure moving 
 
86 LECTURES TO SCIENCE TEACHERS. 
 
 in a plane is known if the motion of any two points (i.e. of a 
 line) in it be known. The motion of any body having 
 conplane motion is known if the motion of a plane section 
 of it, parallel to the plane of motion, be known. Such a 
 plane section of it is, of course, simply a plane figure moving 
 in its own plane. The motion of any body having conplane 
 motion (as in nine cases out of ten in machinery), can therefore 
 be determined by the determination of the motion of two 
 points. In speaking now, therefore, of the motion of a line 
 for shortness' sake it must be remembered that we are really 
 covering all cases of conplane motion of solid bodies. 
 
 In Fig. 4 PQ and T? 1 Q 1 are two positions of the same plane 
 figure, or plane section of a body having conplane motion. 
 If now we have two positions (in the same plane) of any 
 plane figure, we know that the figure can always be moved 
 from the one to the other by turning about some point in the 
 plane. The position of the point 0, about which the figure 
 can be turned from the position PQ to the position P L Q 1 
 can be found at once by the intersection of the normal 
 bisectors to PP 1 and QQ V The motion of PQ in the 
 plane is, of course, its motion relatively to the plane, and 
 therefore relatively to any figure (as A B) in the plane. 
 Such a point as we have found here is called a temporary 
 centre, because the turning or motion takes place about it for 
 some finite interval of time. It will be remembered that not 
 only the two points and PQ of the figure, but every other 
 point of it, must have a movement about this same point 
 at the same time. Now suppose we have some further posi- 
 tion of the same figure, as for example at the position marked 
 P 2 Q 2 , we can find in the same way the centre about which the 
 figure must be turned to move from P 1 Q 1 to P 2 Q 2 . We may 
 indicate this point as r Similarly taking other positions of 
 this figure P^Q S and so on, we can find other points, 2 S , &c. 
 By joining the points OOfl^D^. we obtain a polygon, and if 
 the figure in its motion come back to its original position the 
 polygon also comes back on itself, and passes again through 
 the point 0. Such a polygon, whether it be closed in this way 
 or not, is called a central polygon ; its corners are the tem- 
 porary centres of the motion of the figure. 
 
 I have pointed out that all the points in the figure PQ move 
 round during the motion from PQ to P^. They move 
 round necessarily through some particular angle, the angle 
 
THE KINEMATICS OF MACHINERY. 
 
 87 
 
 POP V and every point moves through the same angle, which 
 we may call r As the figure may have any form- we choose, 
 let us suppose it so extended as to contain a line which is the 
 same length as 00 V and which makes with 00 1 the angle <p v 
 that is to say, the angle through which the figure moves about 
 0. Such a line is shown in Fig. 4 by 
 
 FIG. 4. 
 
 We have, then, a line MM l forming a part of the figure 
 PQ, equal in length to 00 1} the points and M coinciding, 
 and the angle O l MM l being = ^> 1 . Then when the figure has 
 completed its motion about 0, MM l and 00 1 must coincide. 
 Take further similarly M^M^ X 2 and so placed that when 
 
 M l coincides with O v /. O^M^M^ = fa, then when the figure 
 takes its third position, completing the turning about O v M l 
 J/o coincides with 0^^. Similarly we can obtain MJ\f v &c. 
 
88 LECTURES TO SCIENCE TEACHERS. 
 
 The figure thus found is another polygon, which we may call 
 a second central polygon. 
 
 These polygons have important properties, the principal of 
 which can be very easily recognized. The first polygon does 
 not alter its position during the motion of the body ; it is 
 therefore fixed, so that it may be considered as a part of 
 any figure such as AB which is fixed or stationary in the plane 
 of motion. The second polygon moves with PQ and forms 
 (by construction) part of the same figure with PQ. This 
 second polygon then, by the consecutive turnings of its 
 corners upon the corresponding corners of the first (and 
 equal-sided) polygon, will give to PQ the required changes 
 of position relatively to the fixed plane or to the figure 
 AB lying in it. 
 
 If therefore we know the central polygons for the given 
 motion, we know not only the changes of position of the 
 points P and Q, but those of every other point connected with 
 the moving figure, whatever form it may have. For at any 
 one instant every point in the figure is moving about the 
 same centre. In studying the relative motions of the figures 
 we may, therefore, quite leave out of sight their form if ws 
 only know the. central polygons for the motion. These teil 
 us, so far, all about the motion which is taking place. 
 
 We may go further, however. We have recognised the 
 fact that the relative motion of two figures or bodies may take 
 place equally whether one or the other of them, be fixed, or 
 both moving. In the case before us we have supposed AB fixed 
 and PQ moving relatively to it. The second polygon then 
 moves on the first, and expresses the relative motion taking 
 place. If, however, we suppose PQ fixed and AB moving, 
 then the polygons still express, the relative motion ; but the 
 second is now fixed and the first rolls upon it. This follows 
 directly from the constitution of the polygons. The properties 
 of the polygons as expressing the relative motions of the bodies 
 to which they belong are therefore reciprocal. 
 
 You will have noticed, no doubt, that the polygons do not 
 express continuous motion. They define only a series of 
 changes of position in their beginning and end, not telling us 
 of the intermediate stages. 
 
 We may, however, take the consecutive position of the 
 figures as close together as we like. The closer together they 
 are taken the shorter become the sides of the polygons. If 
 
THE KINEMATICS OF MACHINERY. 
 
 89 
 
 at last the distances PP V P^P^ QQ V &c., be taken infinitely 
 small, each corner of the polygon will be infinitely clo*e to 
 the next one. That is to say the two polygons will become 
 curves, and of these curves " infinitely small parts of equal 
 length continually fall together after infinitely small rotations 
 about their end points." In other words the two curves 
 roll on one another during the continuous alterations in the 
 relative position of the two figures. Instead of finding 
 points now by the intersection of normal bisectors, they 
 are found by intersection of normals to the paths of P and Q 
 
 Fio. 5. 
 
 (Fig. 5). The turning about each point now occurs not 
 (in general) for a finite period, but for an instant only. 
 Each point is therefore called an instantaneous centre. The 
 curve containing all the instantaneous centres, or the Locus 
 of instantaneous centres, is called a centroid. Without giving 
 them any special name, several writers on Mechanics have 
 made more or less use of these curves. Among these I 
 may mention Dwelshauvers-Dery, Schell and Prbll. Eeuleaux 
 has, however, given them a name (Polbahnen), and has made 
 
90 LECTURES TO SCIENCE TE AGREES. 
 
 some special use of them, more, I think, than has been made 
 by former writers. 
 
 While the polygons only represent a series of isolated 
 positions of a body, the centroids, rolling on each other, repre- 
 sent the whole motion continuously. Like the central poly- 
 gons their properties are reciprocal. If then the centroids of 
 two figures be known, their relative motions for a series 
 of changes of position, each infinitely small, are also known, 
 i.e. their motions are completely determined. 
 
 If A and its centroid be fixed, and the centroid of P Q 
 rolled upon it (Fig. 5), we have now the means of determin- 
 ing the path of motion of every point in the Fig. P Q 
 relative to A B, whatever may be the form of P Q. It is 
 
 sometimes of great convenience to be able to find the motions 
 of all points in a body in such a very simple way. Eecipro- 
 cally we can determine the point paths of A B relatively to 
 P Q, which, in general, differ entirely from those of P Q 
 relatively to A B. 
 
 If both figures be moving, as frequently happens in practice, 
 both centroids are also in motion ; their motion relative to 
 each other, however, remains unaltered. They still roll on 
 one another, and their point of contact is still the instan- 
 taneous centre of the motion of each relatively to the other. 
 Each figure moves, relatively to the other, about this point, 
 which, being common to the two centroids, is common to the 
 two figures. They might, therefore, for the instant, be 
 
KINEMATICS OF MACHINERY. 91 
 
 connected a that point by a cylindric pair of elements. 
 There are many problems of which the solution is greatly 
 simplified by the recollection of this fact. The point in each 
 figure which coincides with the instantaneous centre, has, 
 therefore, no motion relatively to the other figure. We have 
 already seen this in the special case where the one figure is 
 stationary, for then the point in which the moving centroid 
 touches the fixed one is, by hypothesis, also stationary for 
 the instant ; in other words, it has no motion relatively to the 
 fixed centroid. We now see the general condition of which 
 this is a special case. 
 
 Fig. 6 shows the centroids for the higher pair of elements 
 of Fig. 1. The curve-triangle U TQis the centroid of the 
 
 Fio. 7. 
 
 triangle ABC, and the shaded duangle P V Q W is the 
 centroid of the duangle R P S Q. 1 As the duangle moves 
 in the triangle (the elements sliding upon each other), its 
 centroid rolls within the centroid of the triangle. Both 
 centroids are in this case formed of arcs of circles, and all 
 the point-paths (being determined by the rolling of one 
 circular arc upon another) are combinations of trochoidal 
 arcs. 
 
 The centroids of kinematic chains are generally of greater 
 complexity that those of the pairs of elements just mentioned, 
 
 1 These centroids are shown on a larger scale, apart from the elements 
 to which they belong, in Fig. 2. 
 
92 LECTURES TO SCIENCE TEACHERS. 
 
 but in some cases are quite as simple. In Fig. 7, for example, 
 is shown a mechanism familiar to engineers, in which a crank 
 a drives a reciprocating bar c by means of a block b working 
 in a slot. The centroids denning the relative motions of the 
 links a and c are the two circles shown in full lines, one 
 double the diameter of the other. These two circles both 
 move as the mechanism works (supposing the link d to be 
 fixed), but always so that they roll continuously one on the 
 other. If instead of fixing d the crank a were made the 
 fixed link, the same centroids would still express the relative 
 motions of a and c. The smaller circle, the centroid of a, 
 would be stationary along with the link to which it belongs, 
 and the other would roll on it, the instantaneous centre for 
 the motion of the link c being always at their points of 
 contact. This mechanism (a being fixed) is used in Oldham's 
 coupling, in elliptic chucks, &c. Knowing these centroids 
 we know all about the motions of the two corresponding 
 links in the mechanism, not only about the motions of some 
 particular points in these links. 
 
 The centroids of kinematic chains can in general be very 
 easily determined. Once found they make us independent to 
 a great extent of trigonometric or algebraic formulae, and 
 enable us to determine all we wish to know by purely 
 geometric graphic constructions. For technical purposes at 
 least this is frequently an immense advantage. There are very 
 few cases in which it is not more convenient for the engineer 
 to employ a construction than a formula, if both give him 
 the same result. 
 
 Before looking at the centroids of other mechanisms it is 
 necessary to examine one particular case which often occurs. 
 Suppose that the lines P Pi and Q Q 1 in Fig. 4, or the 
 tangents to the curves at P and Q in Fig. 5, had been parallel. 
 It is obvious that the normal bisectors in the one case and the 
 normals to the curve in the other then become also parallel, or, 
 as it is for some reasons more convenient to express it, would 
 meet at an infinite distance. The temporary centre in the 
 one case and the instantaneous centre in the other are at 
 infinity. A centroid may therefore contain one or mere points 
 at an infinite distance, may have, that is, one or more infinite 
 branches. This constantly occurs in mechanism, and in some 
 cases every point in the centroid is at an infinite distance. 
 This is however a special case ; its treatment does not offer 
 
THE KINEMATICS OF MACHINERY. 93 
 
 any practical difficulty, but I cannot do more than mention 
 its existence here. 
 
 The centroids of the connecting rod and frame of the 
 ordinary steam-engine driving mechanism (the links b and d 
 
 FIG. 8. 
 
 of Fig. 8) may serve as an illustration of this. When the 
 crank a is at right angles to d, the normals to the paths of the 
 two points 2 and "3 are parallel. The instantaneous centre of 
 b relatively to d is therefore at an infinite distance. Each 
 centroid has, therefore, a pair of infinite branches. 
 
 We may look, in conclusion, at one other case which 
 possesses some special interest on account of the form taken 
 by the centroids. It is shown in Fig. 9. The chain con- 
 tains four links and four parallel cylinder pairs. The alter- 
 nate links are equal, and the two longer links are crossed so 
 that the chain forms an " anti-parallelogram " in every position, 
 the angle at 2 being always equal to that at 4, and the angle 
 at 1 to that at 3. If the link d be fixed, the links a and 
 c become two cranks which revolve in opposite directions 
 with a varying velocity-ratio. The centroids of b and d are 
 a pair of hyperbola having their foci at 2 3 and 1 4 respec- 
 tively. The one rolls upon the other as b moves, the instan- 
 taneous centre in the position shown being at the point of 
 contact 0, which is the point of intersection of 1 2 . and 
 3 4. The centroids of the two shorter links are the two 
 ellipses which are shown in dotted lines. They are confocal 
 with the hyperbolae, and their point of contact is always at 
 the intersection of 1 4 and 2 3. Their form shows at once 
 that the rotation of the axes 1 and 4 is precisely the same as 
 that which would be communicated by a pair of elliptic spur- 
 wheels having the centroids for their pitch ellipses. 
 
 In this mechanism, as in some of the others illustrated, the 
 centroids of two adjacent links, as a and d, or b and c, are simply 
 
94 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 a pair of coincident points which roll upon each other. They 
 form thus a limiting case of centroids, but every theorem 
 
 which applies to the more extended centroidal curves applies 
 also to these points, as can easily be seen on examination. 
 
THE KINEMATICS OF MACHINERY. 95 
 
 THE KINEMATICS OF MACHINERY. 
 
 LECTUKE II. 
 
 WE shall in the present Lecture examine in some detail a few 
 of the results which can be obtained by treating mechanisms 
 upon the plan which Rouleaux has proposed, and which is 
 illustrated by his models ; that is to say, by the analytical 
 treatment of which we have already seen the general nature. 
 
 We have seen how kinematic chains are built up from pairs 
 of elements and links. The pairing and the linkage renders 
 the relative motions in the chain absolutely determinate, and 
 the determinate relative motion exists equally whether or not 
 any link of the chain be fixed relatively to the earth or to any 
 portion of space that we choose to treat as stationary. 
 
 We have now to consider in more detail the effect of fixing 
 one link of the chain. In practice, of course, one link is 
 always fixed, or in other words, its motion relatively to the 
 earth, to a locomotive or whatever it may be, is made zero. A 
 chain with one link fixed is simply what we know as a me- 
 chanism. 
 
 In examining pairs of elements we saw that we could fix 
 either element of the pair with lower pairs, the relative motions 
 remaining unaltered ; with the higher pairs the inversion gives 
 us a totally different motion. We have seen also that we can 
 fix any one link of a kinematic chain just as we can fix either 
 element of a pair. We therefore can get as many mechanisms 
 from any chain as it has links. From any such chain as Fig. 
 3, for instance, which has four links, we can get four mecha- 
 nisms. The fact that a kinematic chain gives us as many 
 mechanisms as it has links appears, looked at from this 
 point of view, a mere matter of course. It has, however, 
 never been hitherto distinctly recognized, so far as I know, 
 and it can hardly be realized too distinctly, the consequences 
 which result from it being most important, as we shall see. All 
 that I shall attempt to do in this lecture will be to look at 
 some of the mechanisms obtained from the particular chain 
 just mentioned, and various modifications of it. 
 
96 LECTURES TO SCIENCE TEACHERS. 
 
 We have already noticed that the chain has four links. We 
 see further that it is a chain in which all the motions are con- 
 
 Fio. 10. 
 
 plane, each of its four pairs being simply a cylinder pair, and 
 the four cylinder pairs having parallel axes. It is so propor- 
 tioned that by causing one link to swing, another one can be 
 made to revolve. In order that we may refer more easily to 
 the links a letter is attached to each in the engraving. 
 
 For convenience sake we may also use a short symbol for 
 this chain (the one used by Reuleaux) namely, ((/i) 1 The (7 4 
 within brackets stands for the four cylinder pairs, the symbol 
 for parallel being added to indicate their relative positions. 
 This is the symbol for the chain, no link being fixed. To 
 distinguish the four mechanisms formed from it, we shall put 
 the letter which stands for the fixed link in the position of 
 an index after the formula. Thus we can denote the parti- 
 cular mechanism shown in Fig. 10, in which the link d is 
 fixed, by the formula (Cl) d . We have here then the first of 
 the four mechanisms we can get from this chain. You will 
 recognize it easily enough as exactly similar to the beam and 
 crank of a beam-engine. The link c is half the beam, a the 
 crank and b the connecting-rod. The whole mechanism is an 
 excellent illustration of what I said in my last lecture, that the 
 form of the links is indifferent. If you think of the mecha- 
 nism as forming part of a beam-engine, for instance, you will 
 see in the link d the abstract form of what is generally a most 
 complex structure, a bed-plate with its bearings, an entabla- 
 ture and plummer-block, cast-iron columns, and in some cases 
 even brick and masonry. All these are represented by the 
 fixed link d so far as their kinematic relations are concerned. 
 
 If now we fix the connecting-rod b instead of fixing the 
 
 1 In words " C parallel 4," 
 
THE KINEMATICS OF MACHINERY. 
 
 97 
 
 link d as before, we have the mechanism^) 6 . It does 
 not essentially differ from (Cl) d . .<- The crank now revolves 
 about the pin 2 wh&h was formerly the crank-pin, and the 
 pin 1, which formerly represented the crank-shaft, is now 
 the crank-pin, but there is nothing changed in the nature of 
 the mechanism. By this inversion therefore we have got 
 nothing new. 
 
 Let us now fix the link a, which was formerly the crank 
 (Fig. 11). We have now the mechanism (CQ*; it contains 
 
 FIG. 11. 
 
 precisely the same elements as before, and the relative motions 
 of the links are unaltered, but as a mechanism it is entirely 
 different. It is now a combination frequently enough used 
 in mills and' elsewhere, known by the name of a " drag-link 
 coupling." The links b and d have become cranks, and one 
 drives the other by means of the link c. 
 
 By fixing the remaining link of the chain, the link c, 
 
98 LECTURES TO SCIENCE TEACHERS. 
 
 (which we have supposed to be longer than a), we have the 
 entirely different mechanism (C" 4 ) e , (Fig. 12). The two arms 
 no longer revolve but only swing, and the link a turns right 
 
 round once in every double swing of d and 6. This mechanism 
 is occasionally used in part of its stroke in parallel motions 
 with some modifications, but is not so well known as the 
 others. 
 
 The four inversions of this one chain therefore give us 
 three different mechanisms. Looked at separately it is hard 
 to see the relation in which these stand to each other ; from 
 the point of view which we have taken their mutual relation- 
 ship has become at once evident. 
 
 Without altering the pairing at all, we can greatly alter 
 the chain by changing the relative length of its links. If 
 we made all the links equal we should have a square, of 
 which all four inversions would give similar mechanisms. 
 If we make 6=d and c=a we get a mechanism which is 
 perfectly familiar in the couplings of locomotives and many 
 other cases. All four mechanisms are again similar, each 
 one consisting of "a pair of cranks revolving with equal 
 velocities and connected ]by a link which moves always 
 parallel to itself. 
 
 These mechanisms are among those which have the 
 peculiarity to which I alluded yesterday, that in one of their 
 positions their motions are not determinate. This occurs at 
 
THE KINEMATICS OF MACHINERY. 99 
 
 the " dead points " when a and c are both standing in the 
 direction of the axis of d or b. If no means be taken to 
 prevent it, it is then possible to move the crank either in 
 one direction or the other, and the two cranks may go on 
 revolving in the same direction, or may revolve in opposite 
 directions according to circumstances. 
 
 Such an indeterminateness is, of course, inadmissible in 
 machinery, where we therefore adopt the well-known method 
 of combining two mechanisms of the same kind, and placing 
 them with their cranks at right angles, so that they do not 
 cross the dead-points at the same time. The motions are 
 thus made determinate and the cranks revolve in similar 
 directions. We might, however, wish them to revolve in 
 opposite directions, as in the mechanism shown in Fig. 9. 
 It may be worth our while to look for a moment at the means 
 which may be used in this case to secure the determinateness 
 of the motions in the mechanism. To distinguish between 
 the two cases we may call the former " parallel cranks " and 
 represent it by the formula (Cl \\ C'), and the latter "anti- 
 parallel cranks/' (C" Z ~Z.C"^). This chain, with the link d 
 fixed, is shown in Fig. 13. 
 
 In the case of the parallel cranks all points of the 
 centroids of b and d are at infinity, for they are at the inter- 
 sections of the parallel links a and c. We have already 
 seen, however, that in the mechanism Fig. 9 the centroids 
 are quite different, those of b and d being hyperbolae. If, 
 therefore, when the mechanism is brought into either dead 
 point, where the cranks might change from the anti-parallel 
 to the parallel position, we can only make certain that 
 the right centroids roll upon, each other, we shall get the 
 
 FIG. 13. 
 
 motion that we want., Fig. 13 shows an arrangement by 
 which, just in that position of possible change, a tooth made 
 
 E 2 
 
100 LECTURES TO SCIENCE TEACHERS. 
 
 on one link and a recess upon the other link gear together at a 
 point corresponding to the point of contact of the centroids. 
 The teeth G and H are virtually formed upon the centroid 
 of b, and the recesses E and F upon that of d. At the 
 points where these come into gear, the two centroids are com- 
 pelled to roll upon one another, just as the pitch circles of 
 two toothed wheels are compelled to roll on one another, and 
 in this way the mechanism is carried over its only indeter- 
 minate point, and the cranks remain continuously anti- 
 parallel and revolve in opposite directions. 
 
 This antiparallel chain gives us two different mechanisms. 
 Fig. 13 shows us (C'^Cy. In the other mechanism 
 (CJ2!^s)* ^ e ^ wo cran ks revolve in the same direction with 
 very varying velocity ratios. 
 
 Returning again to the chain (4), it will be seen at once 
 that we may substitute for the pair of elements at 4 a slot 
 and a sector concentric with it, as in Fig. 14. The motions 
 
 FIG. 14. 
 
 remain entirely unaltered. By adopting this construction, 
 however, it becomes possible to construct the mechanism 
 without covering with it the centre of the pair 4, i.e., the 
 point of intersection of the links c and d. "W e can therefore 
 lengthen these links without making the mechanism incon- 
 veniently large. The only constructive alteration is that 
 the slot becomes flatter as the links are lengthened. If 
 we lengthen them little by little until they become infinitely 
 long, the curved slot becomes straight, and its centre line will 
 pass through the point 1. The mechanism modified in this 
 fashion takes the extremely familiar form, already shown in 
 Fig. 8. It now contains three cylinder pairs with parallel 
 axes; the fourth cylinder pair has become a straight slot 
 with a block working in it, namely, a prism pair (see page 
 
THE KINEMATICS OF MACHINERY. 101 
 
 80). The axis of the prism pair is normal to the axis of the 
 three cylinder pairs, and we may therefore use the symbol 
 ((7s P^-) for the chain in its new form. There are here 
 again four links, and therefore four inversions, and we shall 
 find that all four mechanisms are now different. 
 
 We have first the mechanism shown in Fig. 8, and familiar 
 by its continual use in direct-acting engines (pi P- 1 -) 4 . Next, 
 following the same order as before, we may fix the link 6, 
 the connecting rod of Fig. 8. The mechanism thus obtained, 
 (C" P^-f, is quite different from the former, but equally 
 
 FIG. 15. ' , , ' . 1 
 
 * > ' 
 
 familiar (Fig. 15). To make it more recognisable, the prism 
 pair 4 is reversed in the figure, that is the link c is made to 
 carry the open prism and d the full one. The motion is 
 obviously unaffected by the change. The mechanism can be 
 easily seen to be that of the oscillating engine. The link c 
 corresponds to the cylinder, swinging on fixed trunnions at 3, 
 and the link d to the piston-rod and piston of the steam-engine. 
 We see then that the relation between the mechanisms 
 which are familiar to us as the driving-trains of the direct- 
 acting and oscillating engines, is simply that they are differ- 
 ent inversions of one and the same chain. 
 
 Let us now suppose the chain fixed upon the link which 
 was the crank in the last two mechanisms (Fig. 16). This 
 gives us a third mechanism which entirely differs from either 
 of the two former ones. It is quite familiar as a " quick- 
 return " motion in some machine tools, for which purpose 
 also the mechanism last mentioned has sometimes been 
 used. 
 
 Fixing, lastly, the link c, we get the less familiar mechanism 
 shown in Fig. 17 (C" z P-L) . The link b swings about 3, and 
 the crank a rotates in space somewhat as in the mechanism 
 
102 LECTURES TO SCIENCE TEACHERS. 
 
 (Cl), which we have already seen. This train has some 
 practical applications in machinery, hut is not very often 
 used. 
 
 Here then we have ohtained from one and the same chain 
 
 four entirely different mechanisms, all of them more or less 
 familiar. The method we have adopted has again heen 
 successful in making the real relations of these apparently 
 dissimilar things perfectly ohvious. 
 
THE KINEMATICS OF MACHINERY. 103 
 
 We Lave seen in connection with Fig. 1 4 that we can to a 
 certain extent alter the size and extent of a pair of elements 
 without altering its nature or changing the motion of the 
 chain to which it belongs. This alteration in the size of 
 elements, or what may be called the " expansion " of elements, 
 is a process continually carried out by engineers for practical 
 constructive reasons.; and often gives to identical mechanisms 
 extremely different forms. It is impossible here to go into 
 this in detail, a somewhat extreme case of it is shown, for 
 the sake of illustration, in Fig. 18. Here we have the 
 mechanism shown already in Fig. 8, ((7 3 ' J P- L ) <i . The pin 
 of the pair 2 is so enlarged as to include altogether the pair 
 3, the connecting-rod 6 being simply a circular disc, with 
 an eccentric cylindric hole in it. The pin 1 again (the 
 " crank-shaft "), is made large enough to include the whole 
 of 2. We have therefore 3 within 2, and 2 within 1. 
 We have one very common illustration of the extent 
 to which this expansion of pairs is carried practically 
 
 in the link motion. The curved link and block are in their 
 kinematic relations simply a very much expanded pair of 
 cylindric elements reduced in extent by use of a process 
 similar to that by which we got Fig. 14 from Fig. 10. 
 
 We have seen what results have been obtained by making 
 two links of the chain (C^ infinitely long. The same 
 process can be carried still further. In the familiar chain 
 shown in Fig. 7, for instance, three links, d, c and b, 
 are made infinite. We have therefore another prism pair in it, 
 and its formula becomes (C\ 2 P 2 ). It gives us the two 
 mechanisms already mentioned and a third one, all of which 
 are practically applied. 
 
104 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 We must pass over without mention many other modifi- 
 cations and alterations of the chain, and mention only one 
 other form in which it occurs, a form which has some special 
 interest. The condition of movability of a chain that 
 contains four cylinder pairs is not that their axes should be 
 parallel, but that they should meet in one point. The axes 
 are parallel only in the special case where this point is at an 
 infinite distance. Fig. 19 is an illustration of the more 
 general, although less familiar, case when the point of inter- 
 section is at a finite distance. This chain, which may be 
 
 Fio. 18. 
 
 indicated by the formula (CJ-) l has again its four inversions, 
 and furnishes us with three different mechanisms as in the 
 case of (C'). I cannot here go into these ; I mention the 
 chain partly because of its theoretic interest, and partly 
 because although it looks so unfamiliar, it is not unfrequently 
 applied in machinery. If instead of subtending only a small 
 angle, as in Fig. 1 9, three of the links were made to subtend 
 an angle of 90, we should have the common Hooke's or 
 universal joint. In these "conic" chains, links which are 
 quadrants take the place of the infinite links, so that a chain 
 having three quadrants corresponds to the chain of Fig. 7, 
 
 1 "C four oblique." 
 
THE KINEMATICS OF MACHINERY. 105 
 
 in which, there are three infinite links. The formula 
 of the universal joint is (G'jJ- C 1 -)* j ^ corresponds exactly to 
 (6" 2 'PJ-) a and is analogous to (' 4 ') a and (C' 3 '^-L)% Figs. 
 11 and 16 respectively. I have already mentioned in passing 
 that the mechanism (C'^P^) 9 " is that of Oldham's coupling. 
 We see then clearly the close relationship in which this 
 coupling (for parallel shafts) stands to the Hooke's joint 
 or coupling for inclined shafts. It is another illustration 
 of the important results which follow naturally from Eeuleaux's 
 simple method of analysis, and which hardly seem attainable, 
 certainly not with equal directness, "by any other method. 
 
 FIG. 19. 
 
 We may notice one other way in which the chain (C ff 4 ), 
 which we have seen in so many forms, may be modified. In 
 some cases we may not wish to utilize the motion of all the 
 four links, but (say) only three of them ; in that case we may 
 omit the fourth link, if we carry out a proper pairing between 
 the two links thus left unconnected. If, for instance, we 
 omit the link 6 we must make a proper pairing between a 
 and d. This pairing will always be higher ; the chain 
 becomes a reduced chain. Such a chain, reduced by 6, is 
 shown in Fig. 20. The higher pairing is carried out by 
 placing upon a a suitable element, here a circular pin at 2, 
 and giving to c the form of the envelope (see p. 5) of the 
 motion of the pin relatively to it. This envelope is the 
 curved slot shown in the figure. It was not necessary to take 
 the new element at 2, but if it had been in any other 
 place, as 2 1 , the form of the slot would have been more 
 
106 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 complex, as is shown by dotted lines. For such a chain we 
 may use the formula (C" 3 P^) b. The process of reduc- 
 tion can be carried on in this way until only two links are 
 left, which then become really a pair of higher elements. 
 It is constantly employed in machinery, mostly in the case 
 of compound chains, or chains in which some links contain 
 more than two elements. 
 
 The chain which we have been examining has been applied 
 more often than any other to the leading trains of engines 
 and pumps. "We shall in conclusion look at a few of these 
 machines, in order to notice the constructive disguises which 
 often appear in them and render their kinematic identity 
 almost unrecognisable. 
 
 Fio. 20. 
 
 Fig. 21 shows a rotary engine which has been patented 
 several times, and which is founded on the same mechanism, 
 (C" 3 jP- 1 -)*, as the common direct-acting engine. The letters 
 and figures placed upon it correspond to those on Fig. 8, so 
 that the identity of the mechanisms may be the more easily 
 traced. The extraordinary change of form undergone by the 
 connecting-rod b is worth special notice. It has become a 
 bar having a cross-section like a half-moon. It still consists 
 however, kinematically, of two cylinders or portions of them, 
 one described about the axis of 2 (the centre of the disc), 
 the other about the axis of 3. The element 2 is expanded 
 so as to include 1, the crank a becoming therefore a disc. 
 The mechanism is so proportioned that the virtual length of 
 the connecting-rod the distance, that is, between the centres 
 
THE KINEMATICS OF MACHINERY. 
 
 107 
 
 of the elements 2 and 3, is equal to the radius of the disc a. 
 With the mechanism in the form shown in Fig. 21 some 
 separate means has to be provided for keeping 6 against a 
 when the latter is moving upwards. 
 
 Fig. 22 is another rotary engine, which has been patented 
 a dozen times since its first invention in 1 805. It is based 
 upon the mechanism of Fig. 16 (C" 3 P-*-)*. The fixed 
 link a is here made the steam cylinder, while d becomes a 
 
 Fia. 21. 
 
 moving piston. The reference letters are the same as before. 
 It will be noticed that the cylindric element of the link c is 
 expanded to include its prismatic element ; in all the cases 
 formerly noticed the latter had been the larger. 
 
 In order to illustrate this part of his subject Reuleaux 
 examines (in the work I have already mentioned) some forty 
 or fifty rotary engines and pumps all derived form the (G" 4 ) 
 chain and such modifications of it as we have been looking 
 
108 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 at. Models of a number of these are now on the table 
 before you. Many of them bear scarcely any external 
 resemblance to the common steam-engine, to which they are, 
 notwithstanding their dissimilarity, so closely related. "We 
 are not now concerned as to which form is absolutely the 
 best, but are only looking at them from a kinematic point 
 of view. But it is worth noticing that in very many cases 
 not only constructive but mechanical advantages have been 
 claimed for them. Their inventors have over and over again 
 
 FIG. 22. 
 
 claimed some mechanical gain more or less mysterious, com- 
 pared with the ordinary form of engine. Most of them, 
 moreover, have been called "rotary" to distinguish them 
 from reciprocating engines. Our method of analysis, 
 although only kinematic, has shown us not only that there 
 can be no mechanical advantage possessed by one over the 
 other, but that the word " rotary " is essentially a misnomer, 
 if it be supposed to indicate that there is any more or 
 different rotation in them than in ordinary engines. 
 
 We shall now only notice one more of these engines. It 
 is one which has puzzled people a great deal, and with very 
 
THE KINEMATICS OF MACHINERY. 
 
 109 
 
 good reason, for its motions are very strange, and its analysis 
 apparently very complex. 
 
 The engine I refer to is shown (in one form) in Fig. 23. It 
 is known as the " disc engine." It was brought a good deal 
 into notice in 1851, and for a few years was used in the 
 Times office without any ultimate success. Analysis shows 
 that it is based upon the same chain as the Hooke's joint, 
 the conic chain, namely, in which three out of four links 
 subtend right angles, and of which the formula is (0-*-$ C 1 -)*. 
 The fixed link is, however, d, while in Hooke's joint 
 
 FIG. 23. 
 
 a (the acute-angled link) is fixed. Patents have been taken 
 out also for disc-engines in which a and others in which 6 
 is the fixed link. 
 
 I may say, in conclusion, that while it may be impossible 
 for many educational institutions in this country to possess 
 themselves of models so perfect in execution, and therefore 
 so expensive, as those which have been sent to the Loan 
 Collection, it is yet quite practicable to construct many of them 
 in a form which, while quite cheap, is yet well adapted for 
 
310 LECTURES TO SCIENCE TEACHERS. 
 
 educational purposes. I have some made in this way of 
 hard wood with brass pins, which are very serviceable for 
 college purposes. It is just the simpler models, which 
 are the most easily made, which are the most useful in 
 instruction. 
 
 I venture to hope that the . treatment of the theory of 
 mechanism illustrated by the models now in the South 
 Kensington Collection, and of which 1 have here endeavoured 
 to set forth some leading principles, may prove valuable in 
 aiding the study and the comprehension of this branch of 
 machine science both to the student and the engineer. 
 
ON THE STEAM-ENGINE. 
 
 TWO LECTURES. 
 
 BY P. J. BRAMWELL, ESQ., M. INST. C.E., F.R.S. 
 
 LECTUKE L 
 
 AMONG all the subjects to which science lias been applied, 
 it must be universally admitted there is none more tho- 
 roughly and generally important than that of the steam- 
 engine. By the aid of the steam-engine we traverse the sea, 
 by its aid we go at the rate of fifty miles an hour upon rail- 
 ways, by its aid we manufacture every article we wear, we 
 prepare the food that we eat, and we even prepare the ground 
 to receive the grain which we convert into food ; by it we 
 get rid of our sewage, we supply ourselves with water, and, 
 although the creature of fire, by it we extinguish fires ; in 
 fact, it is not too much to say that either directly or indi- 
 rectly, there is no one thing which enters into the ordinary 
 life of a civilised being to the preparation of which the 
 steam-engine has not contributed. 
 
 The subject is so vast that I feel the greatest possible diffi- 
 culty in dealing with it in the course of two lectures ; in 
 fact, so great a difficulty that I hardly know how to begin, 
 or what points to touch. As to the points which must be 
 left untouched, they will of necessity be very many indeed. 
 I think you will be of opinion that (whatever we may be 
 compelled to exclude) it would be impossible, in the neigh- 
 bourhood of this science collection, to refrain from some allu- 
 
112 LECTURES TO SCIENCE TEACHERS. 
 
 sion to the history of the steam-engine, and you will see by 
 the models on the table that that is the conclusion to which 
 I have already come. Therefore I trust you will bear with 
 me while I occupy some portion of your time in going into 
 the historical part of our subject. 
 
 I begin with Papin, who, according to his pamphlet, 
 published in 1707 (in consequence, as Belidor says, of a letter 
 written from London in 1705), states that in 1698, he, Papin, 
 made experiments with his steam-engine, the engine which, 
 no doubt, is so well known to most of you. In the collection 
 there is a cylinder (No. 2007) which is said to be a portion of 
 one of Papin's machines. How that can be, I am at some loss to 
 understand, inasmuch as the cylinder is open at both ends, and 
 at one end only is there a flange or any provision for making a 
 reclosure. The other end, it seems to me, must have always 
 remained entirely open, and therefore I do not know how it 
 could have been employed in any way in Papin's apparatus, of 
 which there is an elementary diagram upon the wall. From 
 this diagram you will see that Papin had invented that which 
 may be called a high-pressure water-raising engine ; but mere 
 water-raising was not the limit (in intention, at least) of 
 Papin's scheme, inasmuch as he proposed an alternative use 
 of the pressure in the air-vessel, viz.: the production of 
 a powerful jet of water from the outlet cock, which jet should 
 impinge on the vanes of a water-wheel, and thereby produce 
 rotary motion. Whether rotary motion was ever so obtained 
 does not appear to be recorded. Although, as I have said, 
 no doubt Papin's engine is well known to most of you, it will 
 be well briefly to describe it. It consisted of a boiler, from 
 which a steam-pipe proceeded to the upper part of a vertical 
 cylindrical pressure-vessel, which could be filled with water 
 from any source of supply above it ; then, the inlet of the 
 water being stopped by the turning of a cock, the steam from 
 the boiler could, by the turning of another cock, be admitted 
 to press upon the water in the vessel and to drive it out- 
 wards through a check- valve into an air-vessel, whence it was 
 to rise to the desired height. To diminish the condensation 
 of steam, a floating diaphragm (which some call and, to my 
 mind, improperly call a piston) was provided to prevent the 
 actual contact of the steam and the water. 
 
 At about the same time Savery, (as we know) in England, 
 was not merely making experiments but was bringing into 
 
THE STEAM-ENGINE. 113 
 
 actual use his water-raising engine. We have a model of 
 that engine here. It comprised the steam-boiler connected 
 "by a pipe (controlled by a cock) with the water-vessel. Steam 
 being admitted and then shut oif, a water-cock was opened, 
 which allowed a cold jet of water to play over the outside of 
 the water-vessel, condensing the steam and making a partial 
 vacuum within the vessel, and thereby enabling it to " suck 
 up " (to use a most convenient but unscientific term) the 
 water from the source of supply ; which water, being retained 
 by a suction-valve, was on the next admission of steam 
 forced through the delivery-valve to the desired elevation, 
 the difference between Savery and Papin being that Savery 
 made use of exhaustion to do part of the raising, and that he 
 did not have a floating diaphragm between the steam and the 
 water to diminish the amount of steam lost by condensation. 
 The engine of Savery was put to work and used to a con- 
 siderable extent, notably at the waterworks, York Buildings, 
 in London, which were situated where Charing Cross Station 
 now is. 
 
 As I have said, Savery's engine did not remain in the 
 region of idea, nor even in that of mere experiment, but 
 it was put to work in several places, and was thus fully 
 tested. These testings, while they showed the merits, showed 
 also the defects of Savery's invention, and these defects were 
 of special importance in days when leaden pipes and plumbers' 
 joints had to be relied on to resist high pressure of steam ; for 
 you will see that in a water-raising engine such as this (as 
 also in the engine of Papin) the steam acting directly upon 
 the surface of the water to be raised must press with a 
 force per square inch equal to the balancing and to the 
 setting into motion (against the resistance of friction as well 
 as that of gravity) of the water to be lifted, and that thus to 
 raise water, say 115 feet, would demand (friction apart) a 
 steam-pressure of 501bs. above atmosphere, or probably, allow- 
 ing for the friction and the loss by condensation, GOlbs. or 
 more in the steam generator. 
 
 With an apparatus such as this, therefore, no alteration 
 either of dimensions or proportions could enable you to evade 
 the fact, that the steam pressure per square inch must be 
 sufficient, not only to balance the column of water, but also 
 sufficient to overcome the friction and all other resistances. 
 Under these circumstances, it is by no means surprising that 
 
 i 
 
114 LECTURES TO SCIENCE TEACHERS. 
 
 the mechanical skill and appliances of those days were 
 unable to cope with the demands made upon them, and 
 that pipes, joints, and cocks, leaked and gave way. 
 
 It was at this time, while the new steam power was struggling 
 into existence, that Newcomen (the father of the modern 
 steam-engine, as I believe him to be) came forward with that 
 invention which in its leading features is used by us to this 
 day. Newcomen said : I will show you a mode of making 
 steam raise water where the pressure of steam per square inch 
 need not bear any relation to the height to which the water 
 has to be lifted ; I will show you that the relation between 
 the resistance and the power necessary to overcome that 
 resistance may be adjusted by causing the steam to act upon 
 some surface not that of the water, and that this surface 
 may be a solid moving freely (but practically steam tight) 
 in a cylinder (this surface being the true piston at last), and 
 that by suitable means I can cause this surface acted upon by 
 the steam to work the other surface whicg is to act upon the 
 water, and that by the due adjustment of the relation be- 
 tween these two surfaces I may make steam of low pressure 
 (or as in practice I intend to effect it, a portion of the 
 pressure of the atmosphere) master of a resistance of hundreds 
 of feet of head of water, and that thus steam barely powerful 
 enough to overcome the atmospheric resistance shall master 
 the resistance involved in the raising of water from out 
 of the deepest mine, or shall produce the high pressure 
 required to distribute water throughout the contracted and 
 tortuous wooden mains for a town supply. 
 
 An engine of this description (a Newcomen engine) was 
 put up in York Buildings to come to the rescue of, and 
 to supersede, the Savery engine which I have previously 
 described. 
 
 Belidor (that most conscientious of writers) states in his 
 Architecture Hydraulique, that he heard of this engine suc- 
 ceeding Savery's at York Buildings, and that he then visited 
 one which was put up at Fresnes ; this he did several times 
 to enable him, as he says, to do justice to it, and to give his 
 readers full information upon it. He has given the fullest 
 possible information, and has left us drawings so complete 
 that any workman in an engineer's factory could at the pre- 
 sent day reproduce from those drawings the very engine that 
 Belidor wrote of in 1737. The leading dimensions were as 
 
THE STEAM-ENGINE. 115 
 
 follows : Cylinder thirty inches diameter, stroke in the cylin- 
 der and in the pump six feet, fifteen strokes per minute, and 
 depth of pit 267 feet. The pumps were seven inches in 
 diameter, twenty-four feet apart, and pumped from cistern to 
 cistern, from the bottom to the top. The boiler was nine 
 feet diameter, three feet deep in the body part, and on the 
 top there was a curious precaution, namely, two feet six 
 inches of masonry to keep that top from being lifted by the 
 pressure of the steam. There were to this engine, gauge- cocks, 
 a safety-valve, and the injection into the cylinder of those 
 clays Newcomen's injection, and a piston kept tight with 
 leather packing with water on the top. The result of the 
 erection of this machine was, that whereas previously fifty 
 horses and twenty men had been employed day and night 
 continuously to keep the colliery free of water, when the 
 engine was completed forty-eight hours of its work unwatered 
 the colliery for a whole week, it requiring only one man 
 on guard to attend to it. Having described this machine 
 in the most detailed manner, Belidor breaks out into a rhap- 
 sody upon it, which being roughly translated is to this effect : 
 "It must be acknowledged that here is the most mar- 
 vellous of machines, and that there is nothing in mechanism 
 which relates so nearly to animal life. Heat is the principle 
 of its movements, there is passing in its different tubes a 
 circulation like that of the blood in the veins ; it has valves 
 which open and shut at the proper times, and it performs all 
 the functions which are necessary to keep it in existence, and 
 to derive from its work all that possibly can be desired." I 
 think those enthusiastic remarks of the excellent writer Beli- 
 dor, in 1737, are as applicable at the present time as they 
 were then, for although the machine is no longer a matter of 
 novelty, and we have got used to it and to its benefits, as we 
 have to those of light and air ; and have thus lost the very 
 appreciation of the existence of its benefits, as we have those 
 derived from light and air, until from some cause they are 
 forced upon our notice. We must, nevertheless, when they 
 are forced upon our notice, agree with Belidor that there is 
 not any machine the product of the intelligence of man 
 which is more worthy of our consideration and commen- 
 dation than is the steam-engine. 
 
 Of the Newcomen engine I have a model here, lent by the 
 Council of King's College. This is such an engine as 
 
 I 2 
 
116 LECTURES TO SCIENCE TEACHERS. 
 
 Belidor describes the boiler with the cylinder above it, the 
 beam, the pumps worked at the other end of the beam, the 
 open-topped cylinder with a piston kept tight by water pack- 
 ing, the water injection into the cylinder, and the plug- tree to 
 make the valve self-acting, and to- cause the machine to con- 
 tinue at work. I have also here, lent by Mr. Bennet Woodcrof t, 
 one of the most interesting engravings to an engineer that 
 I know of. It was published in the year 1719, and shows 
 the ' Newcomen engine erected in 1712 at Dudley Castle. I 
 have also before me another model of a Newcomen engine 
 which is of historical interest. This is a model lent by the 
 University of Glasgow, and bears the label that "in 1765 
 James Watt, in working to repair this model belonging to the 
 Natural Philosophy Class in the University of Glasgow, 
 made the discovery of a separate condenser which has 
 identified his name with the steam-engine." 
 
 To Watt we owe, as you know, the condensation in a 
 separate vessel, the parallel motion, steam-jacket, double- 
 action, expansion, the indicator to tell us how the engine 
 is working, and the governor. We have on the table before 
 us, models brought from the collection, and bearing upon these 
 various subjects. 
 
 Passing on from Watt, we get to Hornblower, who in 1781 
 patented the double-cylinder engine, afterwards largely 
 developed by Wolfe. Leaving the general history and going 
 to a particular class of steam-engines, and going back a little 
 in date, I must refer to Jonathan Hulls, who in 1737 pub- 
 lished the pamphlet I have here, in which he shows how, by 
 the use of the Newcomen engine working a stern wheel, a 
 paddle boat could tow ships out of a harbour against an ad- 
 verse wind, or in the absence of any wind. As the engrav- 
 ing appended to Hulls' pamphlet is not large enough to be 
 distinctly visible, I have had a diagram made of it which is 
 upon the wall. Hulls states that he proposed to employ a 
 Newcomen engine, the rectilinear reciprocating motion of 
 which being obtained by the alternate action of the pressure 
 of the air on the piston for one direction, and of the descent 
 of a counterweight equivalent to half the effect of that 
 pressure for the other direction. Hulls employed this to give 
 alternating reciprocating motion to two pulleys placed upon a 
 shaft, from which by means of bands, one being open and 
 the ot'her crossed, motion was to be transmitted to the 
 
THE STEAM-ENGINE. 117 
 
 paddle-shaft. At each reversal of this shaft, the band which 
 had been driving became slack and ceased to drive, while 
 that which had been slack became tight and did the driving. 
 Obviously, the alternate coming into operation of an open 
 and of a crossed band worked from a shaft driven first one 
 way and then the other, would communicate to the paddle- 
 wheel the desired motion in the one direction. 
 
 Further, on the subject of steam navigation, we have in the 
 collection the actual machine (No. 2150) which, in 1788, 
 worked on Dalswinton Lake in Scotland and drove a small 
 boat. We have likewise the actual engine of Bell which 
 drove the first passenger steamer in Europe. 
 
 Leaving navigation, and going to another branch of the 
 subject, namely, steam as applied to land transit, I must 
 refer you to the extremely interesting model (No. 2145) of 
 Cugnot's steam carriage, which, in the year 1769 traversed 
 the streets of Paris carrying passengers. It went but very 
 slowly, only some two or three miles an hour, and it had 
 frequently to stop to get up steam, but nevertheless it was 
 an actual working common road locomotive, and is therefore 
 a model of great interest, looking at the date at which the 
 original engine was put to work. You will see it is a three 
 wheel carriage, the engine and boiler being attached to the 
 front wheel, so that when the front wheel was slewed by the 
 steersman for the purpose of altering the direction, the whole 
 engine and boiler canted with it. Very curiously \it may 
 be in the recollection of some of you) two or three years 
 since, Mr. Perkins exhibited in the grounds attached to these 
 buildings a common road locomotive, which like Cugnot's was 
 a three wheel machine, and like it, had the engine applied to 
 the castor wheel, and turned boiler and all, with the castor 
 wheel as the steersman moved it. 
 
 In Cugnot's engine we have another instance where rotary 
 motion was wanted, and where there was no provision for 
 obtaining it such as we have now-a-days by means of a crank. 
 Cugnot overcame the difficulty by employing two direct 
 acting inverted cylinders which, by means of palls operating 
 on ratchets, kept the driving wheel revolving always in the 
 desired direction. 
 
 Passing from Cugnot, I now come to Trevithick's engine, a 
 model of which we have here. I was about to ask you to 
 look at the Blenkinsop rail, a sample of which, belonging to 
 
118 LECTURES TO SCIENCE TEACHERS. 
 
 the Institution of Mechanical Engineers, is shown in the 
 Exhibition. It is a rail which, as you know, was provided 
 with teeth projecting from one side of it, into which cogs made 
 in a deep flanche of the bearing-wheel took, so as to ensure that 
 that wheel should move the train and engine and should not 
 slip upon the rail. In Blenkinsop's time this was thought to 
 be a necessity, and such rails were laid down to a considerable 
 extent. I find the sample rail has not been brought in, and 
 I therefore content myself by saying, that a piece of such 
 rail, which was for many years in actual use, is in the 
 collection. In the building also we have the original 
 Puffing Billy, the first engine which traversed, I believe, 
 for practical purposes, a smooth rail, and an engine which 
 worked from 1813 to 1862, and then was placed in its 
 honourable retirement to be revered by engineers. Alongside 
 of that we have Stephenson's "Socket." That engine is gene- 
 rally thought to be in the original condition in which it worked, 
 but a very cursory examination will show that is not so, and 
 those who have read last Saturday's Engineer will have seen 
 a criticism on that Exhibit, pointing out that there have been 
 material alterations, and that therefore it should not be looked 
 upon as showing the exact state in which the engine was put 
 to work in 1830. The other day when calling on Mr. George 
 Robert Stephenson, President of the Institution of Civil 
 Engineers, and now the representative of the Stephenson 
 family, I looked at a model in his possession of the " Rocket," 
 a model which had been made by a man who had driven the 
 engine for many years, and when I found there was doubt 
 cast upon the extent of the alterations, which the real engine 
 had suffered, I thought it would be well to get the model 
 here, and here it is on the table. There is no doubt what- 
 ever that this model corroborates the statement in the 
 Engineer, and shows you really the engine as it worked in 
 1830, without the alterations which have been made in the 
 actual engine now deposited. The alterations, although rather 
 numerous, were not important. In the original engine, as 
 shown in the model, the cylinders were at a considerable 
 angle. You will find that in the real engine they lie at a 
 much less angle. In this model also there is no smoke box, 
 and in the engine there is. I do not know that there are any 
 really important points of difference, but nevertheless when 
 a thing is exhibited for its historical interest, it is as well you 
 
THE STEAM-ENGINE. 119 
 
 should see an exact representation of that which existed at 
 the date under consideration, and I am much obliged to Mr. 
 George Eobert Stephenson for having kindly lent us the 
 model for this afternoon's purpose. 
 
 I have already said, I think, we must look upon it that 
 Newcomen was the first man to make a steam-engine, as 
 we now understand it, an engine with a cylinder and a work- 
 ing piston. Papin, it is true, had a piston, but it was a 
 mere diaphragm acting as a non-conductor between the 
 steam and the surface of the water, and it cannot therefore be 
 regarded as the piston of a working engine. I think we can 
 very readily see why it was that we came to have the New- 
 comen engine in the form of the beam engine originally. The 
 piston was kept tight by water-packing on the surface of it ; 
 if any of the packing leaked through, it did a little harm of 
 course by cooling the steam coming to the cylinder, but not 
 very much, because the piston rose rather by the action of the 
 counterweight than by the pressure of the steam, and any 
 water which leaked through aided the injection in the return 
 (the working) stroke. But to use water-packing it was ne- 
 cessary that the cylinder should be vertical, and that the 
 piston should be pressed downwards in order that the water- 
 packing might lie on the piston. If you press the piston 
 downwards, as in those days was done, and did not work 
 with plunger pumps, but with lift pumps, it involved the use 
 of a beam, which, when the piston went down, pulled the 
 pump upwards, and in this way originated the beam 
 type of steam-engine. JSTow it is extremely difficult in 
 engineering, as in many other matters, when we have 
 once got into the habit of doing a thing, to get rid of that 
 habit. One is apt to take that which is a mere adjunct as a 
 necessity. If you will pardon me, I will tell you, as ger- 
 man to the subject, that which Mr. Eobert Lowe told the 
 Political Economy Club the other night at the Centenary 
 Festival. He quoted the novel of " Hadji Baba," by Morier. 
 where a Persian, visiting England and admiring a young 
 lady, is desirous of making her an offer, and being hospitably 
 received at the house of an English friend, he asked him : 
 " How did you propose to Mrs. Brown 1 " " Well," says the 
 friend, " the fact is, one day we v were going to church to- 
 gether and I offered her my umbrella, and under the shade 
 of that umbrella I proposed to Mrs. Brown." Now the 
 
120 LECTUEES TO SCIENCE TEACHERS. 
 
 Persian made the mistake of supposing that the umbrella 
 was a necessary adjunct to a proposition for matrimony in 
 England, and waiting until he could get a propitiously 
 wet day for the purpose, lost his chance and the young lady. 
 I should have been ashamed to take up your time with 
 telling a tale such as this, had not a grave man like Mr. 
 Lowe thought it not incompatible with his dignity to offer 
 this story to the Political Economy Club as illustrating the 
 mistakes those foreign nations make, who, when wishing to 
 settle a question of tariffs, look upon a treaty of commerce 
 as the vital matter, and not as a mere adjunct, the fact 
 being that the treaty has no more to do with it than the 
 umbrella has in the proposal of marriage. As a fact, 
 however, when your mind is once possessed by an umbrella, 
 or a treaty, or any other thing, it is extremely difficult to 
 get free from it, and therefore when you had once, for any 
 reason, settled on a beam engine it was extremely difficult 
 to get rid of the beam and go to work without it. Besides that, 
 there really were many conveniences attending the use of 
 the beam; the different parts balanced each other, and beams 
 afforded a very convenient means of attaching the pumps, the 
 air-pump and the cold-water-pump and the feed-pump and 
 of making their respective strokes proportionate to the duty 
 required of them. It also, in olden days, was a convenient 
 means of working the plug tree which operated the valves 
 for admitting the steam and for exhausting it. The beam, 
 therefore, had very many conveniences, but we see that when 
 Watt closed the cylinder at the top and made the piston 
 tight without water-packing, one of the very first things that 
 occurred to him was that you might do without a beam at all, and 
 might use the direct action of the piston, by placing the steam 
 cylinder over the pit, and might by an invention of the 
 piston rod, raise the pump without the aid of the beam the 
 construction now so well-known as the Bull engine. But still 
 the beam remained for many years. It had, as I have said, 
 its advantages, and so long as we were using steam for 
 about Slbs. pressure above atmosphere (that was the pres- 
 sure when I was an apprentice), and with some 121bs. 
 vacuum ; and so long as we employed house-engines, the 
 beam being supported on girders built into the engine-house 
 wall (either into the side walls, or in the case of a pumping 
 engine into the end wall, the house only covering half of 
 
THE STEAM-ENGINE. 121 
 
 the engine), no great difficulties arose. But looking on this 
 beautiful model of a six-column engine (familiarly known as 
 a " bed-post engine "), the favourite type of some thirty 
 years ago for factory purposes, you will see that such a ma- 
 chine, when subjected to rapid alternations of strain, due to 
 the use of high-pressure steam and of great expansion (on 
 which important points I hope to address you in my next 
 lecture), would be little competent to bear such strains j for 
 a machine to support them must have its parts well tied to- 
 gether, and our American friends, in the beam engines which 
 they have used with so much success in river steamers, have 
 appreciated this fact, and therefore instead of a " bed-post " 
 frame support of this kind, their walking beam, as they call it, 
 is generally at the apex of a triangular structure of very great 
 strength and of large base, but nevertheless even the American 
 frames are very cumbrous, and one can see how desirable it is 
 to arrive at a more simple form of engine. Notwithstanding 
 this desirability, the beam remained, and when we came to 
 paddle-engines we still used the beam, although we inverted it. 
 I have here a model which is really of historical interest, lent 
 by Messrs. Maudslay, of the engines of the " Great Western," 
 the first paddle steamer built for the purpose of crossing the 
 Atlantic, though I do not say it was the first which actually 
 crossed it. Here is the model of the inverted beam engine of 
 forty years ago, and I am sorry to say that the engineers of those 
 days cared more that the framing should accurately represent a 
 gothic window or an Egyptian temple, or some other piece of 
 architecture, than that it should be designed to resist the strains 
 which came upon it, and as a fact, they made an elaborately 
 ornamental but weak frame which had to depend on the 
 strength of enormous keelson, in the bottom of the ship to 
 keep the parts of the engine themselves in position. Notwith- 
 standing the strength of the keelsons, one frequently found the 
 frames broken and patched with wrought iron. 
 
 By slow degrees we have, however, come to the direct- 
 acting engine the Bull engine, which I have shown you, is an 
 early example of it. Then we had some forty years ago or more 
 the horizontal engine, made by Taylor and Martineau, that 
 type of engine which is now probably more commonly used 
 than any other. Then we had by the elder Brunei an engine 
 with the two cylinders inclined at an angle of 45 degrees each 
 to the horizon, and by their connecting rods laying hold of 
 
122 LECTURES TO SCIENCE TEACHERS. 
 
 one crank pin, the cylinders being placed on an A frame ; the 
 construction known for many years as the Thames Tunnel 
 engine. Then we had by Maudslay (and used in the Eichmond 
 steamboat, I think, as long ago as 1827) the oscillating engine, 
 a most compact form, and one which has been largely employed 
 for marine purposes. Here we have a model of that oscillating 
 engine applied " Thames Tunnel " fashion, that is, the two 
 cylinders at an angle of about 45 degrees to the vertical, and 
 their piston rods laying hold of a single crank. We have 
 then in the way of direct engines, the direct overhead engine, 
 where you have a vertical cylinder, and a connecting rod 
 leading up to the crank \ and those engines, or a pair of them, 
 are much used now-a-days as winding engines in collieries, the 
 drum or a pair of drums between the two cylinders making 
 the fly-wheel. Then we have the same kind inverted with a 
 cylinder carried on side frames, and the connecting rod working 
 a crank shaft near the level of the ground, a construction 
 known as the " steam hammer " and largely used in screw 
 steamers. Further illustrative of this construction we have a 
 wooden model of the compound cylinder engine by Messrs. 
 Rennie of the Peninsular and Oriental steamer "Pera," a 
 form of engine which admits of very ready bracing to with- 
 stand the strains which come upon the various parts. 
 
 The locomotive, of course, is a grand example of the direct- 
 acting horizontal engine, and so are most of our agricultural 
 engines. Besides this, we have had the trunk engine which 
 has been very largely used, either in the single trunk or in 
 the double trunk form (a trunk coming out at both ends of the 
 cylinder), where the piston body and the trunks working 
 through stuffing boxes are the guides to take the inclined 
 thrust of the connecting rod. There are doubtless some other 
 instances of direct-acting engines which do not occur to me at 
 the moment. 
 
 Except for the purpose of pumping and for hammering 
 iron, and for matters of that kind, we almost always want 
 the power developed by a steam-engine to be delivered in the 
 form of rotary motion, and one would naturally say, " Why 
 do we make these machines, beginning with reciprocating and 
 ending with rotary iriotion ? why do we not obtain rotary 
 motion to commence with ? " I need not tell gentlemen here 
 that this has been the problem of the mechanician from the 
 time of W att to the present day. I do not propose to enter 
 
THE STEAM-ENGINE. 123 
 
 into hardly any of the schemes devised for this purpose, "but as 
 you know, for some reason or another, it has "been found not 
 desirable up to the present time at least (for I am sure I should 
 be sorry to limit invention) to employ engines of that kind. 
 At least so far as I know there is not one in practical work ; 
 although there have been some extremely ingenious sug- 
 gestions. The Earl of Dundonald (Lord Cochrane) made a 
 celebrated rotary engine at the place where I was apprenticed. 
 It worked either with air or with steam, and so long as we 
 worked it by exhausting air and letting the atmosphere press 
 upon it, it ran satisfactorily, but when we put steam into it, 
 and it had to support the variations caused by expansion, 
 there were always difficulties. Nevertheless Lord Cochrane 
 was a bold man, and went on. We made a locomotive for 
 the Greenwich Eailway, he would design his own boiler, and 
 that boiler never raised steam to speak of ; and the only result 
 of the steam that was raised was, I am sorry to say, the de- 
 struction of the engine. Earl Dundonald, good sailor as he 
 was, had forgotten his anchorage, he had only anchored his 
 engine by the steam pipe, and when the steam was put on, 
 the engine preferred to twist the steam pipe round instead of 
 causing the driving wheels to revolve, and the unfortunate 
 locomotive was consigned to the scrap heap. There is at the 
 present time under trial a rotary engine of the most extraordi- 
 narily simple character. It is the invention of a Scotchman 
 long resident in America, Mr. Dudgeon, who may be well 
 known to many of you as the inventor of the "tube expander." 
 This engine consists of nothing at all but two spur wheels with 
 teeth that work steam tight in each other, and at their ends 
 work steam tight also against two plates. Assuming the spur 
 wheels to be on horizontal axes placed side by side, and that 
 it is intended they should revolve so that the teeth which are 
 in contact move upwards, then the steam would be intro- 
 duced through one of the end plates a little above the line of 
 centres ; it would there be received into the cavity between 
 two pairs of teeth, and it would press upon the under teeth 
 to drive them down, while it would press upon the upper 
 teeth to drive them up. It might be thought that under these 
 circumstances no motion would ensue ; but- it will be seen 
 that the area of the upper surfaces would be larger than the 
 area of the lower, and thus there would be a preponderat- 
 ing upward effect. After the teeth have passed a particular 
 
124 LECTURES TO SCIENCE TEACHERS. 
 
 point they act as a slide valve and cut off the steam, and the 
 steam contained between two teeth goes on expanding until 
 a further point is reached where by their divergence thev 
 allow it to escape. I have seen the engine at work, and most 
 certainly 60 Ibs. of steam went into the engine and the ex- 
 haust steam came out perfectly inert. My notion was, that 
 the 60 Ibs. steam was being utilised. I have endeavoured to 
 get results from trials by the dynamometer, but those trials 
 were inconclusive. I merely bring this engine before you as 
 one of the ingenious attempts what its future may be I 
 know not to obtain that which we all feel ought to be 
 obtained, the production of rotary motion from steam at first 
 hand instead of the obtaining it at second hand through 
 reciprocating motion. I may say that this science collection 
 offers very little inducement indeed for me to speak to you 
 about rotary engines, for so far as I know it contains 
 but three specimens. Here, in one by the late Patrick Bell, 
 we have a cylinder bent into a semi-ring, with a con- 
 tinuous slot on its inner side ; a disc, working steam-tight in 
 this slot, carries pistons which you will see are intended to 
 enter and to leave the cylinder edgeways, but while in the 
 cylinder to lie flatways, so as to be pressed upon by the 
 steam. I do not know that this engine was ever more than an 
 idea. I do not know the date of it, and I do not think any of 
 us will look at it as likely to give economical results. Here is 
 another machine which has been commonly classed as a rotary 
 engine, but I am not sure that it is correctly .so classed ; it is 
 the celebrated Disc engine, from which so much was expected 
 at one time. It is in truth a reciprocating engine because the 
 disc reciprocates from side to side, changing the point of 
 contact, and by means of a rod standing out at one of its faces 
 gives rotary motion to the crank. After all there is a crank. 
 The crank is a necessity, and therefore such engines can 
 hardly be classed among the rotaries. The third model I 
 hava here is little more than a Barker's mill. 
 
 There is no doubt that many of the persons who set them- 
 selves to devise rotary engines, do so, because they are under 
 the impression that there is an absolute loss of power by the 
 use of the crank. I need not tell gentlemen who are here 
 to-day, science teachers, that there is no loss of power at all 
 theoretically speaking. As a matter of fact you may by the 
 friction of the shafting and its bearings, by the necessity of 
 
THE STEAM-ENGINE. 125 
 
 having a fly-wheel and matters of that kind, turn a certain 
 amount of power into heat in your bearings, which had better 
 be employed in driving the machinery, and therefore and also 
 for the purpose of simplicity, if it could be attained, the rotary 
 engine is a thing to be desired. But, commonly speaking, the 
 efforts to obtain a simple engine, end in disappointment, and 
 in much greater complication than prevails in the machines 
 which they are intended to supersede. 
 
 Having made these few remarks on the history of our sub- 
 ject (which remarks have carried me much further than I had 
 intended), I will now ask you to consider with me what are 
 the points to be borne in mind in constructing an economical 
 steam-engine. In the expression " steam-engine," I think I 
 must (although want of time tempts me to do otherwise) 
 include the boiler, that very essential element. Were it not 
 so essential, I would ask you to suppose that by some means 
 we had obtained steam economically, and that the object of 
 the Lecture was to consider how far that steam so obtained 
 could be used with advantage. But I think probably it will 
 be as well that we should devote some little time to the boiler 
 itself. As to the kind of boiler to be used, the first thing to 
 be considered is safety ; the next things in very many in- 
 stances indeed are weight and compactness. When we have, 
 for example, to consider the boiler of a common road locomo- 
 tive like Hancock's, of which I have a diagram on the wall, 
 or when we have to consider the boiler of a locomotive or 
 the boiler of an ocean steamer occupying space which might 
 be filled by cargo paying freight, or when we have to consider 
 boilers put into buildings in the city of London where land 
 lets for a guinea a square foot per annum, we may readily see 
 that the compactness of the boiler is an element which weighs 
 very materially in its construction. There are, however, many 
 cases for other purposes, such as water-works, large cotton- 
 mills, and establishments of that kind where the engine con- 
 structor is at liberty to give the boiler any size and weight 
 he thinks necessary, so that occasionally we have to take 
 into consideration the question of compactness, and occasion- 
 ally it is a matter of comparative indifference. In old days 
 (which, I am sorry to say, means when I was an apprentice) 
 the kind of boiler commonly in use for land engines was 
 the waggon boiler, a boiler set over the fire, the products of 
 which went under the boiler bottom to the end, and origi- 
 
126 LECTURES TO SCIENCE TEACHERS. 
 
 nally did not return through a central tube, but returned by 
 a flue passing along one side across the end, and away by 
 the other side up the chimney. An improvement on that 
 construction was to make a central tube to add to the 
 surface, and when that was introduced, the products of com- 
 bustion came back along the central tube, then divided and 
 went half on one side and half on the other, by the side 
 flues to the chimney at the back. These waggon boilers were 
 largely used in cases where steam was employed, up to about 
 6 Ibs. pressure, but were not employed for pressure much 
 exceeding this. You will see from the diagram on the wall that 
 the waggon boiler is one of an extremely weak form. There 
 were occasionally stay rods connecting the crown of the arch 
 with the sides, but after all it was, as I have said, an ex- 
 tremely weak form, and was not competent to bear more than 
 the 31bs. to 6 Ibs. pressure which came upon it. Then our 
 Cornish friends showed us how to make boilers with the fire 
 contained within them the boiler which we English engineers 
 look upon as a source of economy ; but we must not authori- 
 tatively pronounce it to be so, because the French, who are 
 a highly scientific nation, dispute it altogether,- and declare 
 that you get such imperfect combustion in the case where 
 the fuel is surrounded by water that it is better to revert to the 
 system of having the fire below the boiler, coupled with the 
 use of lower tubes, " boilers " as they call them ; this favourite 
 French construction of boiler is known by us as the elephant 
 boiler. However, there is no doubt that the English continue 
 to look upon the internal fire as the correct thing, and I may 
 say that boiler insurance engineers also look upon it as being 
 far preferable to the impinging of heat externally upon a boiler 
 shell ; in fact, I think I am right in saying that the gentle- 
 man who stands at the head of that branch of our profession, 
 lays it down as a maxim that there ought not to be direct 
 fire heat applied to any portion of the boiler subjected to 
 internal bursting pressure. There is no doubt whatever 
 that within the last year and a half there have been two 
 instances of fatal explosions of boilers composed of a number 
 of tubes of not more than about 12 inches in diameter, com- 
 petent to bear a cold-water pressure of several hundred 
 pounds on the square inch. One of these explosions was 
 at a pressure, I believe, of only 50 or 60 Ibs. These 
 boilers were insured by an association I have in my mind ; 
 
THE STEAM-ENGINE. 127 
 
 an explosion occurred, and the engineer at the head of that 
 association is now, as I have said, convinced that it is 
 undesirable ever to expose any part of a boiler subjected 
 to a bursting strain (a strain from within tending to rend 
 it) to fire impinging upon it. 
 
 I will therefore go back to the Cornish boiler, that is to 
 say, to the cylindrical boiler with a single cylindrical internal 
 tube, in one end of which is made the fire, and through the 
 other end of which go the products of combustion. Such a 
 boiler as that contains plenty of water ; there is ample space 
 for the water to circulate, and also a large surface which is 
 free to absorb the heat, and there is ample- surface of water 
 to give off the steam, which is a matter to which too much 
 attention cannot be paid. As an illustration of this subject 
 of the requisite surface for the delivery of the steam, we all 
 know that when we take the cork out of a bottle of soda- 
 water, full nearly to the neck, the gas will rush out and 
 drive the water before it. If we half empty the bottle, 
 recork it, and stand it vertically and let the gas accumulate 
 and then again take out the cork, you will, see a violent 
 agitation, but if you incline it so as to increase the surface of 
 the water and take out the cork there is much less agi- 
 tation. The extra surface has insured tranquillity, as a large 
 surface in a boiler insures tranquil delivery of the steam. 
 The fact being that as the particles of steam in a boiler 
 have to escape at the surface of the water, unless that 
 surface is adequate in proportion to the amount of steam 
 to be generated it is impossible to obtain a quiet delivery 
 of the steam, and in lieu thereof you have a violent 
 ebullition, and you have particles of water carried over 
 with the steam, mixed with it in the form of fine rain; to 
 which we give the name of priming. I may here, perhaps, 
 be asked by you how is it that in a locomotive boiler, which 
 is no larger than one of these Cornish boilers, but which, as 
 we all know, has so much greater evaporating power, we are 
 able to get dry steam at all. The answer to it is this. The 
 locomotive is worked at a very much greater pressure than 
 that at which most of these Cornish boilers work, and the 
 increase of pressure tends to cause tranquillity, and does so for 
 this reason : if I deliver in a given time 1 Ib. weight of steam 
 from the surface of the water at atmospheric -density I must 
 deliver a certain number of certain sized bubbles of steam 
 
128 LECTURES TO SCIENCE TEACHERS. 
 
 in this time. If I deliver in the same time 8 Ibs. weight of 
 steam from the same surface of water under a pressure of eight 
 atmospheres, those bubbles would be only as numerous and as 
 large as were the bubbles of the 1 Ib. of steam escaping from 
 the water at the pressure of one atmosphere, because the steam 
 being eight times as dense it is obvious that you get it off in 
 a more compact form, and that, therefore, the mechanical 
 agitation of the water is no greater under those circumstances 
 than it would be with the lower pressure and the smaller 
 quantity of steam, and that this is so is practically known to 
 every intelligent man who has the conduct of a steam-engine. 
 Such a man is well aware that if a boiler primes, the way in 
 which he can check it is by partially closing the regulator 
 so as to keep the pressure up. We equally know that if the 
 pressure suddenly falls in a boiler from any cause, such as 
 the blowing away of a safety-valve, every drop of water conies 
 out of the boiler with the rush of steam, the fact being, 
 that tranquillity at the surface, which tranquillity is necessary 
 to prevent priming, is to be obtained by proportioning the 
 ratio of the water surface to the number and size of the 
 bubbles of steam (no matter what their density) which have 
 to come through it in a given time. 
 
 The foregoing has been a digression. I will now ask you to 
 consider the question of boiler surface. In the diagram A 
 on the wall with its single flue representing the Cornish 
 boiler, let us consider for the present the flue surface only. 
 Compare with it diagram B where the single flue is 
 replaced by seven flues each J the diameter of the large 
 one in A. It is perfectly certain, that by that means we 
 have 2J times the surface in B we had in A. Then you 
 will see if you divide each one of the seven tubes into 
 seven again, you get 2J times 2J which equals 5|- the 
 original surface. In that way you might say to me, if 
 you were not engineers, why not carry through to infinity 
 and get an infinite surface in the boiler ? There is a very 
 good answer to it ; in fact several answers, any one of which 
 is good. One is that if you have too much surface in the 
 boiler in proportion to the steam to be delivered, you do not 
 have a sufficiency of water between the tubes to insure that 
 there shall always be water in contact with them. Another 
 is that you do not afford a passage between the tubes for the 
 steam generated by the lower tubes to get clearly away to the 
 
THE STEAM-ENGINE. 129 
 
 water surface. Another is that when the water is foul, the 
 tubes become so coated with deposit as to fill up the spaces 
 between them almost to a solid, and you cannot clean them. 
 But another and a very important reason is this through 
 these tubes have to go the products of combustion. Those 
 products commonly have not completed their combustion 
 in the fire-box, but you require it to be completed in the 
 tubes ; and if they are too small you have made your boiler 
 into the meshes of a Davy lamp, and by means of these 
 tubes you have as effectually snuffed out and extinguished 
 the gases which go through them, as the meshes of a Davy 
 lamp extinguish the inflammable gases which endeavour 
 to pass through them. You therefore have very good reasons, 
 all of them, why the multitubular system should not be 
 carried too far. In practice in steam-boats the tubes are 
 generally from 2J to 3J inches in diameter, with probably 
 from J to 1J inches of water space and with a length of tube 
 of 6 ft. 6 in. In locomotives, where you have a forced draft and 
 therefore do not require so large an area to get the products of 
 combustion through, the tubes average probably from 1J to 
 If inches diameter, and their length varies from 8 to 10 feet. 
 The tubes of the portable agricultural engine lie between 
 these dimensions. To-morrow I hope to say a word to you 
 about the proportion of surface, but in speaking of the actual 
 proportion I am inclined to state that, as far as my experience 
 goes, it matters very little indeed, except upon the scores of 
 safety and ability to clean, and of having a large water 
 surface to deliver steam, what kind of boiler you employ, 
 so long as it has an adequate absorbing surface for the heat 
 evolved from the fuel in a given time. I believe the various 
 contrivances or notions about vertical surfaces and flat surfaces, 
 and all these things which have agitated engineers' minds for 
 many years, are absolutely unimportant so long as you have 
 an adequate absorbing surface for the heat produced. It is much 
 more important to have a construction which will afford dry 
 steam, power of cleaning, and of repair. I will now ask you to 
 consider a little how much of a successful economic result is 
 due to the proportioning of the boilers, and to good stoking. 
 On the wall is a table, which is only a small portion of one 
 published by the Eoyal Agricultural Society relating to the 
 performance of portable engines under trial by that 'Society. 
 You will find two columns showing the number of pounds of 
 
130 LECTURES TO SCIENCE TEACHERS. 
 
 water actually evaporated by 1 Ib. of Llangennech coal, 
 assuming the water to be at 212 which it was very nearly. 
 The other column shows the water that could have been 
 evaporated had the feed water been at the ordinary tem- 
 perature of 62. We will confine our attention to the first 
 column, and you will see there that the average obtained was 
 9-85 Ibs. of water boiled off by 1 Ib. of coal. This column 
 is sufficient to show the extreme danger of judging by an 
 average unless you know the maximum and the minimum, 
 for if you look at the second line from the top you will find 
 that as much as 11*83 was boiled off in one case, while in 
 another case it was only 4-84. The same coal was used ; the 
 difference was in the proportions of the boiler, and in the 
 ability of the stoker. With respect to the stoking on these 
 trial occasions, the stoking is* done by men, some of whom 
 are paid several hundreds a year. It is done by a hand 
 fire-shovel, more like a banker's scoop than an ordinary 
 shovel; the coal is broken into bits about the size of a 
 walnut, the fire-door is opened about once in- every two 
 minutes (which you will find very wrong, according to 
 Tredgold, and so it would be if the men were not trained to 
 do it by sleight of hand so dexterous that it is hardly possible 
 to see the operation performed) ; and in that way there is 
 kept up an absolute uniformity of combustion, a uniformity so 
 great that, during the five hours which the trial sometimes 
 lasts, I will undertake to say the pressure gauge does not vary 
 one-sixth of a pound. So accurate is the firing that whereas 
 the engine would blow off with 80 Ibs. and thus waste steam 
 which would spoil all the competitor's chance, at 77 J Ibs. 
 the engine could not run against the weight, therefore the 
 stoker's margin of possible working in these engines lies 
 within 2J Ibs. Such accuracy you cannot hope to obtain in 
 daily work. Let us see how essential good stoking is. What 
 is it that we want to do 1 We want the most perfect com- 
 bustion of the fuel, coupled with the least admixture of air, 
 for the purpose of making that combustion. Assume for the 
 sake of illustration that every pound of coal gives us 14,000 
 English heat units if we burn it to carbonic acid, and that 
 we only get in round numbers 4,000 units if we burn it 
 to carbonic oxide, that is to say, in the latter case we only 
 get -3^- and lose ^ of the effect. What we have to insure, 
 therefore, is, that sufficient air is introduced into the furnace 
 
 
 
THE STEAM-ENGINE. 131 
 
 to consume the whole of the fuel and convert it into carbonic 
 acid, and that none for want of air shall escape out of the 
 fire in the condition of carbonic oxide, but, on the other 
 hand, every particle of air which enters the fire-grate beyond 
 the quantity required is a source of loss by increasing the 
 weight of heated material which escapes unutilized up the 
 chimney. It is a common thing for newspaper writers to 
 point out that smoke is not only a nuisance, but is a waste 
 of fuel which might be burnt. So it is, but it is far less 
 waste than would go on if you had furnaces so managed 
 as to pour up the chimney carbonic oxide. This being in- 
 visible does not excite attention, and thus two men might be 
 employing fuel ; one sending up no visible smoke (although 
 plenty of carbonic oxide), would receive commendation as 
 an economical man, while the other, who did make smoke 
 (which he ought not to do), but made no carbonic oxide, 
 would be improperly stigmatised as the extravagant man. 
 I believe what we have to do is to see if we can obtain 
 mechanical stoking, and on this point I should like to 
 be allowed to say a few words on combustion generally. It 
 appears to me that you may use fuel either in the form 
 of gas, as is practised by Dr. Siemens ; in the form of 
 liquid, as in the use of hydrocarbons ; in the form of almost 
 impalpable powder as used by Mr. Crampton, or in the 
 ordinary form in which we get it as coal, and as we employ 
 it in steam-engines. If you employ it as coal, then clearly 
 the only portions which are available for combustion at 
 one time are the surfaces. In order to develop sufficient 
 heat for your purpose, you are compelled to have within the 
 fire-box, or on the fire-grate, as much coal or coke as will 
 when you take its surface into account (if you could measure 
 it), unite with enough oxygen to produce the required heat in 
 a given time ; and for this to be done you must either have 
 a Jarge extent of grate, or must keep the fire open, or must 
 have a powerful draft ; with a large extent of grate, there 
 is the difficulty that it should at all times be covered, 
 otherwise air runs freely through the spaces uncovered by 
 fuel, causing much loss. On the otljer hand, with a great 
 depth of fire you have the very difficulty I have been point- 
 ing out that of making carbonic oxide. Carbonic acid is 
 produced in the lower part of the fire, and this carbonic 
 acid rising through the hot fuel above, takes up another 
 
 K 2 
 
132 LECTURES TO SCIENCE TEACHERS. 
 
 atom of carbon, and goes out as carbonic oxide. There are 
 these great difficulties in actual manual firing, and it is to 
 get over them that the engine-driver at the Koyal Agri- 
 cultural trials fires once in every two minutes, and obtains 
 practical uniformity from the beginning to the end of the 
 live hours' trial. But, as I say, this is impossible in practice. 
 When fuel is dealt with in other ways it is far more 
 manageable. It is all surface, if I may say so, when it is 
 reduced to impalpable powder, because the surface of a given 
 weight in that form may be popularly said to be infinite as 
 compared with the surface of lumps of coal, and therefore 
 when working on this system you can properly proportion the 
 supply of powder and of air to each other, and so obtain ex- 
 tremely good results by this mode of firing. At present I 
 am sorry to say that Mr. Crampton is not helping us in the 
 economic generation of steam, for he is devoting himself to 
 the application of his system not to the obtaining heat for 
 steam-engines but to the production of the high temperatures 
 required in the puddling of iron. In the case of the liquid 
 hydrocarbons, and in that where the fuel is converted into 
 gas, as in Dr. Siemens' producers, it is also possible in prac- 
 tice to regulate with nicety the amount of the air, and it is 
 from this power of regulation that the large percentage of use- 
 ful effect has been got from liquid fuel, and that the high 
 results arising from the combustion of gas have been got even 
 when that combustion is unaccompanied by the regenerative 
 system of Dr. Siemens, which system has caused the vast 
 economy of fuel which has been obtained since its introduc- 
 tion in many of our most important metallurgical and other 
 processes. 
 
ON THE STEAM-ENGINE. 
 LECTURE II. 
 
 REVERTING to the subject of boilers, I did not yesterday 
 mention the kind of boiler which is in most general use 
 now-a-days in England for manufacturing purposes I 
 mean the Lancashire boiler, the internally fired two-flued boiler. 
 It probably is one of the very best that we have from a 
 practical point of view, as it gives in a comparatively small 
 diameter a large area of fire grate, and it gives also the con- 
 siderable heating surface of the internal tubes. Moreover 
 it affords ample opportunity for getting in below the tubes 
 for the purposes of cleaning or of repairs when necessary. 
 The marine boiler of the present day is, as we know, a boiler 
 from 11 to 12 or 13 feet in diameter, made of plates up 
 to an inch in thickness, and competent to stand a pressure 
 of steam of 70 Ibs. above atmosphere. It contains generally 
 three cylindrical fire-tubes, the flame-box, the return-tubec, 
 and a dry uptake. The marine boiler of old days was, in 
 truth, nothing but a tank, the form of which was determined 
 by consideration as to its adaptability to the shape of the 
 vessel more than by any other rule. The flues were rect- 
 angular and flat-sided, and it was wholly unfit for use except 
 with low pressure. To show you how low the pressure was, 
 I may state that there was more pressure on the bottom of 
 such a boiler (considered merely as a tank) from the weight of 
 water in it (when the steam was not up) than was pressing 
 against the under side of the top of the boiler when steam 
 was up. 
 
 The cylindrical cases of boilers resist the tendency of the 
 internal pressures to burst them by means of the tensile 
 strength of the plates and of the riveted joints ; and flat 
 surfaces, where they must exist, such as at the ends of 
 cylindrical boilers, are stayed either by stay-bars or by 
 gussets ; or, in the case of the fire-boxes of locomotives, 
 by screw stays 4 inches apart and upwards ; but there is 
 
134 LECTURES TO SCIENCE TEACHERS. 
 
 another set of circumstances to be considered, namely, that 
 when cylindrical flues are exposed to external pressure and 
 I think we owe it to the late Sir William Fairbairn that 
 that which was formerly a most fruitful source of giving way 
 in steam-boilers has been cured, because, although a perfect 
 cylinder subjected to pressure ought to bear more than it 
 would now, the pressure applied as an internal bursting strain, 
 yet even a perfect cylinder when pressed from the outside is 
 in a state of unstable equilibrium, and, in practice, a flue thus 
 pressed, therefore, is subject to being destroyed at almost 
 any moment, either by its not having been originally a 
 perfect cylinder, or by differences of temperature, or by 
 wear ; and we know as a fact that such flues collapsed 
 under pressure, and thus they were a constant source of 
 accident in internally fired boilers. Now this defect has been 
 entirely cured by the expedient of riveting at intervals T 
 iron, or a specially made sort of channel iron, round about 
 the flues, but there is one very simple mode applicable to 
 all existing boilers which I think is equally efficacious with 
 the riveting system, and is better in other respects, as there 
 are no rivets to be acted upon by the fire. That mode 
 consists in simply cutting out of a piece of boiler-plate a 
 ring (which if it be in an existing boiler may be made in 
 halves and bolted together), then if that ring be slid on over 
 the tube so as to fit it, the edge of the ring where it touches 
 being bevelled off, in order that there may be as little a part 
 of the tube cut off from water contact as possible, it will be 
 found, although there is no attachment between the ring and 
 the tube, that you have most effectually prevented the tube 
 from collapsing, for the simple reason that it cannot diminish 
 its diameter in one direction without correspondingly increas- 
 ing it in another. In that way without any rivet or attachment, 
 it is possible to make flues of Lancashire and Cornish boilers 
 safe against collapsing. 
 
 The only other kind of boiler I will mention to you is one 
 which was in use a great many years ago, and was called the 
 " Quicksilver-boiler." It was employed on a large scale (that 
 is to say on a fair working scale), to drive the mill engine of 
 the King and Queen Ironworks, the property of Mr. Howard, 
 the inventor, also on a paddle-steamer called the " Vesta " 
 which ran as a trading boat to Ramsgate. This boiler had 
 a double bottom to it. The upper one (the false bottom) 
 
 
THE STEAM-ENGINE. 135 
 
 having cups projecting downwards to add to the surface, 
 and between the two bottoms, and surrounding the cups 
 there was an amalgam of quicksilver and lead that was kept 
 at a temperature of, I think, about 450. There was no 
 water whatever in the boiler, and when it was desired to 
 start the engine, the attendant, by means of a, hand-pump, 
 injected as much water on to the cup surface as would 
 generate sufficient steam for a half revolution of the engine. 
 The feed-pumps of the engine were forced inwards not by 
 eccentrics, but by springs, and the amount of their inward 
 stroke could thereby be regulated to a nicety by a set screw, 
 and in that way the engine ran without there being at any 
 time an appreciable quantity of water in the boiler, the 
 stored up heat between each injection of water accumulating 
 in the amalgam. This steamboat ran for very many mon-ths, 
 and the factory engine, I believe, for very many years. The 
 government tried it in one of their boats, but eventually it 
 was abandoned, and I merely mention it to you as an interest- 
 ing link in the history of the steam-engine. 
 
 With respect to the staying of boilers, there is one other 
 boiler I must mention, and that is the boiler of Hancock's 
 common-road locomotive to which I alluded yesterday. This 
 was composed of a succession of flat chambers ranged side "by 
 side like books on a shelf, the spaces between the chambers 
 being about one inch, and the chambers themselves being 
 about 1^ inches wide, and, according to the size of the boiler, 
 from. 20 inches to 30 inches square or broad and high.. These 
 chambers thus arranged side by side, were put over the fire 
 which played between them in the spaces. The water was 
 contained in the lower part of the chambers, and the steam at 
 the top, and I think we shall all agree that a more impossible 
 form for withstanding high-pressure steam, it would be diffi- 
 cult to imagine that is, at first sight. But each of these 
 chambers had raised upon it hemispheres, and the hemisphere 
 on the side of one chamber abutted against the corresponding 
 ones on the side of the next chamber, and thus the chambers, 
 say to the number of fourteen, stood side by side, each chamber 
 supporting its neighbour, and there remained only to be stayed 
 the two outside plates of the outside chambers. This was 
 done by means of thick boiler plat3s which pressed on the 
 hemispheres of those outer chambers, and those outer chambers 
 and plates were held together by outer tie-rods and cross 
 
136 LECTURES TO SCIENCE TEACHERS. 
 
 girders. There were also two inner tie-rods, one at the 
 top, and one at the bottom, which, passing through all the 
 chambers and through certain washers, served at the same 
 time as tie-rods, and by the annular spaces between the rods 
 and the insides of the washers as a steam-pipe at the top 
 and a feed-pipe at the bottom. The plates were not above 
 J of an inch thick, and the pressure of steam was, I believe, 
 as much as 80 Ibs. , and those boilers were the means whereby 
 Hancock was enabled upwards of thirty years ago to drive his 
 carriages through the streets of London, and to take passengers 
 from the bank to Paddington and to Whitechapel regularly 
 amongst the omnibus traffic without hindrance of any kind 
 or description. There were various meritorious points in that 
 steam-carriage, which I should like to go into if time permitted. 
 Up to the present it has not only not been surpassed, 
 but I am sorry to say that particular branch of engineering 
 has gone backwards, and up to the present time Hancock's 
 carriage has not been equalled. It was noiseless, smokeless, 
 sightly, easily steered, light and powerful, necessary qualities 
 all for street navigation and for avoiding the alarming of 
 horses. Hancock attended to all these points, while the 
 engineer of the present day, although he may make a very 
 good engine, does not appear to consider that he has to 
 make one which is to go about amongst horse traffic. 
 
 With respect to the proportion of boilers, in the table that 
 I exhibit here of the engines tried at the Royal Agricul- 
 tural Society, I may tell you that the proportion of absorbing 
 surface of the engines varied from 16 superficial feet per 
 cubic foot of water evaporated per hour, to 37 superficial 
 feet. With the 16 feet there was obtained '651 of the total 
 heat effect of the coal, and the gases escaped at 775. With 
 the 37 feet super there was obtained '776 of the total effect 
 of coal, and the gases escaped at 500. I may say in the 
 face of such discrepancies in proportion as those, it is difficult 
 to fix on any rule, but the ordinary rule is from 12 to 15 
 superficial feet of heating surface for each cubic foot of water 
 evaporated per hour, while the surface of grate where there 
 is no forced draft is from -^th to -g^th of the heating 
 surface. 
 
 Before leaving the subject of boilers I should like to call 
 your attention to the solar boiler which is being worked at 
 Tours ; an account of this boiler is to be found in the Revue 
 
THE STEAM-ENGINE. 137 
 
 fndustrielle for the 24th of October last. I dare say the 
 particulars are known to most of you, but as they may not 
 be to all, I will repeat them. The effect of the solar rays on 
 45 feet of surface when concentrated upon the boiler was 
 sufficient to evaporate 11 Ibs. of water in one hour. The 
 construction was simply this. A conical mirror was employed, 
 the sides of the cone being at an angle of 45 ; the diameter 
 of the cone was 2 -6 metres, and the bottom of it had a flat 
 surface of one metre. In the axis of this cone was a copper 
 boiler about 1 1 inches in diameter, and of the height of the 
 cone. The boiler was blackened, and was surrounded by a 
 glass envelope ; there was also an apparatus to cause the axis 
 of the cone to point always towards the sun, and the sun's 
 rays falling upon the sides of the mirror were reflected 
 radially on to the boiler, and the result was, as I have said, 
 to obtain an evaporation of as much as 1 1 Ibs. of water in 
 one hour from the effect of the sun's rays falling on 45 feet 
 super. 
 
 We frequently hear of very wonderful evaporating per- 
 formances, and when they are very wonderful we should do 
 well to distrust them. We must recollect that one of the 
 defects of a boiler is priming, and we must equally recollect 
 that that abstracts water from the boiler, and, therefore, 
 appears to make the boiler do higher evaporating duty, and 
 a bad boiler is thus credited with a merit for that which 
 is a positive defect. I know no means of" making quite 
 sure that this priming does not take place, except by re- 
 condensing the steam that comes from the boiler and observing 
 the heat given forth by it ; but by this precaution it may 
 soon be ascertained whether the product from the boiler 
 has been dry steam, or partially steam and partially water 
 mechanically carried over. 
 
 Assuming that we have once got from some boiler the 
 steam, we will now consider how we are going to use it in 
 the steam-engine, and I will refer you to these two diagrams 
 one where I show by a vertical line the pressure of the 
 steam from zero up to 20 atmospheres, and by the horizontal 
 line the temperatures ; the other (2), showing the volumes 
 corresponding to the various pressures. 
 
138 LECTURES TO SCIENCE TEACHERS. 
 
THE STEAM-ENGINE. 
 
 139 
 
 Now having the steam, there are four modes in which we 
 may use it in an ordinary reciprocating steam-engine. We 
 
 may use it non-condensing and non-expansive ; condensing 
 and non-expansive ; non-condensing and expansive and con- 
 
140 LECTURES TO SCIENCE TEACHEES. 
 
 densing and expansive. I will first of all take the simplest 
 and worst form of steam-engine, that of non-condensing 
 and non-expansive, and I will just ask you to consider 
 how important it is under these circumstances that the 
 steam which is used should be of very high pressure. So 
 obvious is this that it would seem absolutely unnecessary 
 to labour the point, were it not that one knows m practice 
 steam-engine users, and too frequently steam-engine makers, 
 will employ large cylinders to perform a given amount of 
 work in non-condensing, non-expansive engines, having a 
 notion that they are really doing good, and are at the same 
 time acting liberally, whereas they are doing harm and are 
 making the engine wasteful. The effects in a non-condens- 
 ing and non-expansive engine of using different pressures of 
 steam are shown in Diagram 3. I will assume, to begin 
 with, that we have a pressure of steam in the boiler 120 
 Ibs. above zero, and for simplicity and to get rid of fractions I 
 will take the atmosphere as equal to 15 Ibs. At 120 Ibs. 
 above zero if we use a steam-engine without expansion and 
 deduct 15 Ibs. for the resistance of the atmosphere we get 
 from a given quantity of steam a work of 105. If we use 
 it at only 60 Ibs. above atmosphere, deducting 15 Ibs., we 
 obtain a work of 45, while if we use it at 30, and deduct 15, 
 we obtain a work of 15. If, however, we use it at 18 above 
 atmosphere and deduct 15 we obtain a work of only 3 ; 
 that is to say, the percentage of coal required to raise 
 the steam to atmospheric pressure is clearly greater the 
 less is the pressure of the steam. If we try to work a 
 non-condensing engine with steam at atmospheric pressure, 
 evidently, we should get no work out of it at all, although 
 we should expend the coal in boiling off the water ; if the 
 engine were worked at 1 Ib. above the atmosphere, yfths 
 of the coal would go away in boiling the water and Y^-th 
 only would be utilised. But if worked at 120 Ibs. above zero 
 or 105 Ibs. above atmosphere, fth only of coal is lost, and 
 j-ths are utilised. Therefore we see how essential it is that 
 to obtain in a non-condensing, non-expansive engine any- 
 thing approaching to economy, high pressure steam should be 
 employed. 
 
 I will now ask your attention to the advantage in non- 
 condensing engines of the use of expansion. The results 
 are expressed on this diagram, which exhibits the well-known 
 
 
TEE STEAM-ENGINE. 
 
 141 
 
 X 
 
 DIAGRAM 3. 
 
142 LECTURES TO SCIENCE TEACHERS. 
 
 hyperbolic expansion curve of steam, taken on the assumption 
 that the volumes are inversely as the pressures. Supposing 
 we once more begin with the pressure of 120 Ibs. above 
 zero and let the steam in through a space of one, viz., A to a, 
 then shut it off and expand it through an equal space (a to a'), 
 we shall obtain a pressure of 60 Ibs. above zero as the final 
 pressure, and obviously we should have got in addition to 
 the work represented on the upper part of the diagram by the 
 parallelogram (A X a x) of the steam let in, the work repre- 
 sented by the part next below it, this figure (a x a! #'), the 
 area of which is all but -j^-ths of the area of the top figure. 
 Now if one expands that steam which is at this time of the 
 capacity of 2 into the capacity of 4 (a" x" line), it is quite 
 clear that the area contained between the line a x and the 
 a" x" (the steam being now 301bs., above zero pressure 
 finally), is exactly the same as that of the previous figure, 
 because while every one of the horizontal lines, b' c f , &c., is 
 half the length of the horizontal lines, b c corresponding to it 
 in the figure above, every one of the vertical lines is double 
 the length of the similar vertical line above, and therefore the 
 area is the same, and thus by this further expansion we have 
 got nearly a further -j^-ths of the original pressure. Again, 
 if we pursue that and double the volume we get a further 
 i^ths and it would appear on the face of it, as if one might 
 go on ad infinitum and get infinitely great power. That of 
 course cannot be true. It would be at once seen it was not 
 so if, instead of taking a curve varying inversely with the 
 pressures, we had made our lines after allowance for the 
 difference of temperature; then the true expansion curve 
 thus obtained would have been found to be one that could 
 not be infinitely pursued. But even without taking that into 
 consideration it would not pay to carry the expansion curve 
 down to a very great extent, because the size of the machinery 
 would become so enormous ; and therefore, one is compelled 
 for commercial considerations to keep the expansion within 
 certain reasonable limits. 
 
 Diagram 4 shows, in a tabular form, the value of expansion 
 when non-condensing. Supposing in lieu of using a cylinder 
 full of steam at 120 Ibs. the steam is cut off at half stroke, it 
 is quite clear that we have only used half the steam, and the 
 work which we shall get in that part before cutting off will be 
 of the value of 60, but in the part after the cut off we shall get 
 
i 
 
 THE STEAM-ENGINE. 
 
 143 
 
 I 
 
 I 
 
 o 
 
 33 5 g 
 
 ff 
 
 5 
 
 a 
 
 fc 
 <J 
 
 1 
 
144 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 .n 
 
 O CQ 
 CO O 
 
 3 
 
 a 
 
 a 1 
 
THE STEAM-ENGINE. 145 
 
 about Ytfths of that value, or about 41^, mating a total of 
 gross, from which, if we deduct the atmospheric back pressure 
 of 15, we get a work of 86 J, or if we used the whole steam we 
 should have got an effect of 165 per cent, instead of an effect 
 of 100. Diagram 5 shows this further : if we cut off 120 Ibs. 
 above zero of steam, at Jthand employ only 15 parts of steam 
 instead of 120, we get by expansion increments of 10J down 
 to quarter-stroke, 10 ^ again between quarter and half stroke, 
 and 10J again between the half and the bottom, making 46 
 gross, from which we have to deduct the 15 back pressure, 
 leaving 31 as the net work done. Thus ^th of the weight of 
 steam gives 29 1 per cent, of the work that would have been 
 given by the full steam, or if we were to employ full steam 
 we should get 236 of work instead of 100. 
 
 I will now come to the advantage of condensation when not 
 expanding. This is illustrated by Diagrams 6 and 7. Suppose 
 again that we take 120 of steam to work the whole stroke, 
 and that we take non-condensed vapour in the cylinder as 
 equal to 2, we then get 1 1 8 of work done. Now formerly we 
 found that without condensation we should have to deduct 
 the whole atmosphere and get, therefore, only 105 work done. 
 The difference therefore between 105 and 118 is 12 J per cent., 
 the gain one has got by condensation without expansion. 
 That gain of course will become very much greater as the 
 pressures become lower, because with the 60 Ibs. steam, as 
 I showed you, we get only 45 without expansion and con- 
 densation, but we get 58 with condensation, or a gain of 
 28-8 per cent. With 30 Ibs. steam we got only 15, and 
 with condensation we get 28, or a gain of 86 '6 per cent., and 
 with 18 Ibs. steam where we only get 3 non- condensing we 
 now get 16, and, therefore, gain 433 per cent., showing the 
 advantage of condensation applied to a non-expansive engine. 
 
 The last kind of engine we have to consider is the expan- 
 sive condensing engine, and Diagram 8 shows us its advantages. 
 If we use 120 initial pressure, and cut off at a half, the weight 
 of steam used in the cylinder will be 60 ; the useful effect, 
 if non-condensing, would be 86^, and the useful effect con- 
 densing 99 J ; the percentage of gain being 15. If we cut 
 off at a quarter we shall use 30 Ibs. weight of steam, we 
 shall get an effect of 56^ without condensation, or 69 J- with 
 condensation, and the gain will be 23 per cent. If we cut 
 off at Jth we shall use but 15 Ibs. of steam, and we shall 
 
 L 
 
146 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 8 1 
 
 CM 
 i 1 
 
 !l 
 
 
 Hr _ 
 
 o 
 
 
 ^-\ 
 
 c 
 
 04 
 
 
 w 
 
 
 1-1 ei 
 
 
 
 ft 
 
 t 
 
 X X 
 
 
 
 5 
 
 gj 
 
 F-l T i 
 
 
 8 
 
 2 
 
 
 3 
 
 ^ 4 
 
 
 
 HH 3 
 ? 
 
 1-1 rf 
 
 
 
 
STEAM-ENGINE. 
 
 147 
 
 X &3 
 
 O j^ 
 
 . 3 
 
 is, 
 
 O 
 
 PH CC 
 
 S 9 
 
 o s 
 
 s .2 
 
 II 
 
 o o 
 
 X 
 
 g i 
 
 R 
 
 O K 
 . H 
 
 3 a 
 
 |^ ^ 
 
 P O ' 
 
 | 
 
 Z> o 
 
 X o 
 
 2 
 
 8 tS 
 
 ^ 
 
 PJ o 
 
 ^ ^ 
 
 W & 
 
 3 
 
 pa 
 
 E 
 
 S3 . 
 
 -i 2 
 
 <~ % 
 
 o ^ 
 
 g 
 
148 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 ft 
 
 p 
 
 II * 
 
 ""O co 
 
 So 00 
 
 ^3 o 
 
 .s ^ 
 I'l 
 
 P S. 
 
THE STEAM-ENGINE. 149 
 
 still get an effect of 31 without condensation, or with con- 
 densation 44, and the gain per cent, therefore is 42. 
 
 No. 9 is a very similar Diagram, showing you the effect 
 of using 120 Ibs. steam with condensation, but without 
 expansion, and also where you expand that steam twice 4 
 times, 8 times, 16 times, and 32 times, the net useful effects 
 being 118, 199, 277, 352, 418|, 468-f, and the ratio 1, 1-68, 
 2-34, 3, 3-54, 3-89. 
 
 One sees, therefore, that for economy, condensation is a 
 most necessary thing. More especially when the pressures 
 are low, condensation really becomes a duty. There are 
 various means of effecting it. There is the old injection 
 condenser invented by Watt, where the stream of water goes 
 into a chamber and the condensed steam and the liberated air 
 are withdrawn by means of an air-pump. This requires a 
 constant supply of cold water, and that cannot always be 
 obtained in sufficient quantities. Then we have to resort to 
 means of recooling. One of the commonest is, as we know, 
 a cooling pond, but that requires a considerable amount 
 of space. Another one is a pile of brushwood, down which 
 the water is allowed to trickle. In both those cases the loss 
 by evaporation equals the feed water. Another method was 
 that pursued by Howard to be used with his quicksilver boiler. 
 He passed the injection water through pipes which were 
 outside the hull of the vessel, and which were acted upon 
 by the sea-water, and in that way he recooled the injection 
 water and used the same water over and over again. It 
 was a necessity for him to have pure water, because with 
 the boiler I have described a deposit upon the surface above 
 the amalgam if at all serious would very rapidly have caused 
 the boiler to burn out. Then we have the surface condensers, 
 of which we found a model yesterday in Mr. Watt's case, 
 This condenser is largely, in fact almost universally employed 
 now-a-days in marine engines. The proportions of it vary 
 very much according to the rapidity with which the water 
 is driven through, and according also to the temperature at 
 which the water is expected to be found. A steam-boat 
 intended to navigate the Eed Sea should be provided with a 
 much larger surface in her condenser than one intended for 
 the Atlantic. There is a very elaborate paper by a French 
 engineer, an abstract of which has been translated and will 
 be found in the Proceedings of the Civil Engineers, which 
 
no 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 
 
 I H 
 & ^ 
 
 O 
 
 W ^ 
 
 w 
 
 a ^ 
 H 2 
 
 ' ^ 
 6 w 
 
 g g 
 
 s < 
 
 K- I 
 
 ^q 
 
 <5 hH 
 
 CH 
 
 X 
 
 
 05 
 
 K ^ 
 OQ W 
 
 OH 
 
 ^s 
 
 I 
 
 V 
 _ 00 Oi t^ -M 
 
 e3 ^ ^H CM CO 
 
 O <N * 00 CO C* 
 
THE STEAM-ENGINE. 151 
 
 by formulae shows how you may vary the proportion accord- 
 ing to the amount and to the temperature of the water pumped 
 past the cooling surfaces. 
 
 The before-mentioned condensers, whether injection or 
 surface, require a large amount of water, but there is another 
 kind of condenser not much used, but which should be a 
 great deal more used than it is, for the purpose of surface 
 condensing, a condenser by means of which you can, wher- 
 ever you have sufficient water to feed a boiler, efficiently 
 condense the steam of the engine. This condenser consists 
 of a number of rows of horizontal tubes connected at the 
 ends to vertical stand-pipes. Over the topmost of these rows 
 there are troughs, perforated, into which the condensing water 
 is pumped, and from which it trickles over the surface of the 
 condenser and thereby condenses the steam, there being 
 evaporated from the outside of the condenser in the form of 
 mist or reek almost exactly the same weight of water as 
 was used for the feed. The condensing water that is not 
 evaporated is collected at the bottom and is repumped in a 
 warm state on to the top of the condenser, and by that 
 apparatus it is, as I have said, perfectly practicable to main- 
 tain a very excellent vacuum of some twenty-four or twenty- 
 six inches by an expenditure of no more water than that 
 which of necessity must be expended in the feed of a non- 
 condensing engine. 
 
 Besides that we have had condensation effected by means 
 of the air. A good many years ago a gentleman of the name 
 of Craddock invented an air condenser which consisted of 
 a cage placed on the roof of the engine-house and joined 
 by a stuffing-box to the eduction-pipe. This case was caused 
 to revolve in the air, and the contact of the air with the 
 surface effected condensation. I have not the particulars, 
 and I do not think the vacuum obtained was very good, 
 but nevertheless there was a vacuum obtained, and the con- 
 denser was put to work at the London works, Birmingham, 
 then the property of Messrs. Fox, Henderson, and Co. 
 Then for the purpose of bringing the steam back into water 
 to be used over again we have the air condenser of Hancock, 
 who, in his engines, had the condenser made on the same plan 
 as his boiler, through the condenser the air was blown, and 
 that air afterwards fed the fire. 
 
 There are very many benefits attending the use of surface 
 
152 LECTURES TO SCIENCE TEACHERS. 
 
 condensers and evaporative surface condensers (apart from the 
 mere obtaining a vacuum) in substitution of a jet condenser. 
 Among the benefits are, that you get pure water for the boiler, 
 you do not get the boiler fouled by a deposit. On the other 
 hand, you must take care that you do not get the water too 
 pure, because if you have not some kind of scale on the 
 plate, as there is always a certain amount of fat acid which 
 gets into the boiler from the grease, the plates "pit" and 
 rapidly decay. But if you let in just as much new water 
 as is necessary to make up for the waste that inevitably 
 takes place, that, generally speaking, deposits sufficient im- 
 purity to prevent the condensed steam acting on the bare 
 plates. In surface condensation in the marine engine you 
 at once get rid of the necessity for blowing off a certain 
 portion of the heated water to discharge the brine from the 
 boilers; I mean the continuous blow off that was employed 
 when injection condensers were used and where you put 
 into the boiler about double the necessary feed water in 
 order that (the water in the boiler being maintained at twice 
 the saltness, if I may use the term, of that of the sea,) one- 
 half being blown away it may carry with it two particles of 
 salt, the two particles that came in with the double quantity 
 of water. Tliis continuous blow off involved the waste of a 
 great portion of the heat, but this loss has been obviated alto- 
 gether by the use of the surface condenser. Another ad- 
 vantage is, that we get rid of the presence of air in the fresh 
 injection of water that comes in, in the ordinary jet condenser. 
 I have said that in the injection condenser the water is 
 generally abstracted by an air-pump, but we know that it has 
 also been abstracted by means of a pendant pipe of such a 
 length that the pressure of the atmosphere cannot balance 
 the column of water within it, and we also know that for 
 some years past Mr. Morton has employed what is called an 
 ejector condenser, not only for extracting the water and air, 
 but to act as the condenser itself. In engines running at 
 any considerable speed, especially if coupled engines, where 
 there is a sufficiently rapid succession of beats of the exhaust 
 steam, it is possible without the use of any live steam from 
 the boiler and by the means of the exhaust steam, and 
 the jet of water coming in, to cause the products of con- 
 densation to pass out of an ejector condenser against the 
 pressure of the atmosphere, carrying with them the air : one 
 
THE STEAM-ENGINE. 153 
 
 of these condensers was put to work in the building below 
 us about two years ago, applied to a high-pressure engine 
 driving the machinery in the Exhibition. The ejector con- 
 denser is an extremely simple apparatus and one readily 
 applied to any existing engine, and there are in it no 
 working parts. I will not endeavour to describe the principle 
 of its action, because if time permits I propose to repeat 
 the explanation I have given elsewhere of the principle of 
 Giffard's injector, and this explanation is applicable to the 
 ejector condenser. 
 
 About the earliest instance we have of the expansion 
 engine doing admirable duty, is the Cornish pumping-engine, 
 which is shown in diagrams 19 and 19 a. You are aware that 
 those engines are single-acting, that the steam comes in at the 
 top side of the piston, the vacuous condition being below; 
 after the piston has gone down a certain distance, the steam- 
 valve is closed by the action of the plugtree and the rest of 
 the journey is performed by expansion. When applied to a 
 mine the engine lifts the weight of the pump-rods, and on the 
 return journey, the equilibrium-valve being open, the weight 
 of those rods acting on the plunger-pump raises the water 
 up to the surface of the ground. In a mine, therefore, the 
 resistance to this engine is constant, because the depth of 
 the mine being constant the head of water against the 
 pump is constant. Those engines were admirable machines 
 in their day, and they developed an economy which prior 
 to that time had been thought to be impossible. Their per- 
 formance is commonly expressed, as you know, in millions 
 of duty, that is to say, in millions of pounds raised one 
 foot high by the combustion of 1 1 2 pounds of coal. Before 
 the year 1855, the record used to be in millions of pounds 
 raised one foot high by the combustion of a bushel of coal, 
 which was taken as 94 pounds. You are probably aware 
 that in 1811, Mr. Lean, a Cornish engineer, the inventor 
 of the plunger-pump, established Lean's Reporter. He 
 commenced, I think, with three engines in the counties of 
 Cornwall and Devon, and in his Eeporter he gave month 
 by month the performances of the respective engines. A 
 great deal of emulation was caused thereby, and there is no 
 doubt that the establishment of that "Eeporter" was a 
 great incentive to engineers to improve the quality of their 
 engines. The conversion of millions of pounds duty into 
 
154 LECTUEES TO SCIENCE TEACHERS. 
 
 horse-power is one which we ought to have very readily in 
 mind, because we are apt to be led away by the large figures 
 of millions; it is something like a man's income being stated 
 in francs; we want a ready method of converting the francs 
 into sovereigns. 1 have put here a table showing this con- 
 version : one pound of coal per horse-power per hour would 
 equal '22 If millions of duty of the present day. It is easy 
 to remember that one pound of coal per horse power per hour 
 is equal in round numbers to 222 millions of duty, and 
 that 2 pounds of coal per horse-power per hour are equal to 
 111 millions of duty, and so on. I had occasion to speak 
 yesterday of the difficulty even among engineers of getting 
 out of a fashion. The case of the Cornish engine is a further 
 illustration of this; the Cornish engines were very good in 
 Cornish mines, and thereupon Cornish engines were intro- 
 duced into waterworks. Now except for the purpose of 
 pumping up into a reservoir, where the head of water being 
 nearly constant, the engine is in the same condition as it is in 
 when pumping water to the surface of the ground from a 
 mine, I am prepared to give my opinion that the employment 
 of Cornish engines for waterworks purposes is an entire 
 mistake. They operate by the descent of the weight forcing 
 up the water, and that weight being constant must always 
 be balanced by a constant pressure. To obtain this, what 
 the waterworks' engineer does when he employs a Cornish 
 engine to pump into the distributing mains, is to put up a 
 stand-pipe, and if he calculates the maximum resistance the 
 engine should overcome to be equal to a head of 200 feet, he 
 puts up a stand-pipe 200 feet high, and from the top of 
 the stand-pipe he brings a return pipe which goes to the 
 main. Between eight and eleven o'clock in the morning, when 
 so large an amount of water is taken in London from the 
 mains, there is a very great draught upon them, and it is im- 
 possible to maintain 200 feet in them, in fact, there is not 
 maintained above 150 or 140 feet close to the engine-house ; 
 so that for the purpose of using the Cornish engine you pump 
 water up to the maximum height of 200 feet, and let it im- 
 mediately tumble down on the outside uselessly 40, 50, or 60 
 feet. Thus it is that although such an engine may appear to 
 be working economically when the consumption of coal is 
 compared with the stand-pipe lift of 200 feet, it may never- 
 theless be working very wastef ully when that consumption is 
 
THE STEA M-ENGINE. ] 65 
 
 judged of by the 140 or 150 feet of resistance of head actually 
 required ; and moreover there is no necessity for the purposes 
 of economy to have one of these machines, which, in addition 
 to the difficulties I have mentioned as attending its applica- 
 tion to the supplying of distributing mains, has the defect of 
 being unlimited in its range, that is of having no determinate 
 stroke. With a double-acting engine working upon a crank 
 the length of the stroke is determined, and leading into an 
 air vessel or into a single stand-pipe (not a double one, going 
 up and coming down again) such an engine is resisted solely 
 by the load which actually prevails in the mains, and the 
 work that engine is doing is real, whereas a Cornish machine 
 requires a constant maximum pressure, although the mains 
 may not require that constant maximum pressure. I say, 
 therefore, that is an instance of how the engineer is misled 
 by fashion. There are cases where single stand-pipes are used 
 with Cornish engines, and there you will actually see throttle- 
 cocks put on to the foot of the stand-pipe to keep the "head " 
 up, and throttle- valves put in the equilibrium-pipe to prevent 
 these engines from having only their right amount of work 
 to do when that amount of work happens to be too small 
 to balance the working weight. 
 
 To revert to the question of expansion ; in the Cornish 
 engine the steam is cut off by the shutting of a double beat 
 valve (the valve which Harvey and West have introduced 
 also into their pump work), and we have also in slow-going 
 engines valves lifted by cams and dropping at proper 
 intervals. We have then in quick running engines, the 
 Corliss valve moved by an eccentric and brought back to rest 
 by a pneumatic cylinder (which on account of the risk of an 
 air leak, I cannot help thinking is a dangerous thing, I 
 would rather see a steam cylinder than a pneumatic one). 
 The Corliss valves are liberated from the eccentric rod by a 
 spring catch connected with the governor. But one of the 
 commonest modes of making expansion in our ordinary 
 engines, is to have one slide valve upon the back of the 
 other, each valve worked by its own eccentric. By that 
 contrivance it is possible to cut off at almost any part of the 
 stroke. There is no doubt a certain want of sharpness in 
 the cut off, and you do not get on the indicator that beau- 
 tiful-looking figure which you do by the Corliss valve, or the 
 drop valve of any kind, but when one comes to work it out, 
 
156 LECTURES TO SCIENCE TEACHERS. 
 
 the loss by the imperfection of the outline of the figure is 
 extremely small. There is also the expansion used in the 
 locomotive by the link motion. I perhaps ought not to dwell 
 upon that now, for if time should permit 1 may say a few 
 words about the locomotive engine. 
 
 The objection to all expansion in single cylinder engines is 
 the variation in the load and strain throughout the stroke, 
 and the necessity of making the cylinders of great size, and 
 also the proportioning of parts so as to bear the initial strain, 
 although the average pressure is small when the expansion is 
 high. To diminish those difficulties double cylinder engines 
 (the Hornblower or Wolfe engines) were invented, and are 
 now largely employed. Of these there have been several 
 constructions. A common one with beam-engines is to put 
 the high-pressure cylinder nearer to the centre of the beam 
 than is the position in which the low-pressure one is placed, 
 and by the shorter stroke thus obtained and by a diminished 
 diameter of bore the desired proportion of contents of the 
 two cylinders is obtained. We have had also cylinders placed 
 side by side at the end of the beam ; we have had the 
 high-pressure cylinder placed within the low pressure, and 
 their piston rods coupled up to one cross-head. In this case the 
 internal cylinder should always be steam-jacketted, but I do 
 not look upon this arrangement as a very happy one. 
 
 We have also had the cylinders placed end on by putting 
 the high-pressure cylinder below the low-pressure and letting 
 the piston-rod come through. Great difficulties have been 
 experienced in arrangements of this kind by having in- 
 accessible stuffing-boxes, and so on, but these have been 
 cured in later constructions. Then we have had McNaught's 
 system, which possesses great advantages for the con- 
 version of non-expansive old-fashioned beam-engines into 
 expansive engines by the application of a second cylinder. 
 McNaught puts his expansive cylinder at that end of the 
 engine which is between the beam centre and the crank shaft, 
 and thereby instead of taking the strains of the two cylinders 
 through the beam-gudgeons and the bolts, the strain of the 
 high-pressure cylinder is brought on the crank side ; in that 
 way an engine which would not stand the application of the 
 high-pressure cylinder put in the ordinary position, will 
 support that application very well. These are the modes in 
 which compound cylinders have been employed. There may 
 
THE STEAM-ENGINE. 157 
 
 be others which do not occur to me at the moment. Consider 
 ing the principles of the double-cylinder engine it is readily 
 seen that if the steam were simply used throughout the full 
 stroke in the small cylinder and were then turned into the 
 large one, there would not be any reduction of initial strain 
 whatever on the piston of the large cylinder nor on the 
 parts attached thereto, because if there were no loss of pres- 
 sure, although there would be a constant decrease as the 
 steam expanded, yet, at the moment of turning it on there 
 would be the same pressure on the low-pressure piston as if 
 you turned the steam in directly from the boiler. But now- 
 a-days we expand in the high-pressure cylinder to a greater or 
 less degree, and thus the piston of the low-pressure cylinder 
 does receive much less strain than if the boiler steam were turned 
 upon it. This arrangement tends to diminish the pressure on 
 the low-pressure cylinder, and it tends also to equalize the 
 strain, producing a more equal tangential force, but these ad- 
 vantages are obtained at a very considerable cost. I am sorry 
 I have not got a diagram here, which would have shown you 
 practically why it is when you have a compound cylinder en- 
 gine you obtain no more work out of the two cylinders than 
 you would obtain if you put the same quantity of steam into 
 the low-pressure cylinder alone, and gave it the same amount 
 of expansion. I do not say the high-pressure cylinder is 
 not doing work, because it is, but that which it does is 
 abstracted from the low pressure cylinder, and in that way, 
 therefore, one has all the complication and expense of the 
 two cylinders, and the final result got out is no more than 
 if one had a single cylinder of the same size as the low- 
 pressure cylinder not as the . aggregate size of the two 
 and expanded the steam in that cylinder to just the same 
 extent. We have thus a very large amount of compli- 
 cation introduced into an engine for apparently a very 
 small end. However, it is so introduced, and it is the or- 
 dinary construction now-a-days in marine steam-engines. It is 
 said there is a utility in it, because one not only diminishes 
 the strain on the reciprocating parts, but the tangential forces 
 are equalised to a great extent. That- is true, but I want 
 you to see what the value of that is. I have here Diagram 10, 
 which shows by Fig. 1 an engine working without any ex- 
 pansion, and by the figure below it the tangential effect in 
 producing the motion on a crank. 
 
158 LECTURES TO SCIENCE TEACHERS. 
 
 DIAGRAM 10. 
 
THE STEAM-ENGINE. 159 
 
 Obviously, the non-expansive engine has a tendency to 
 produce variation of motion, that is to say, the tangential 
 force will be zero when the crank is on the centre, and it will 
 be at the maximum when the crank is at half-stroke. Jn 
 constructing these diagrams I assumed the connecting rods 
 to be of infinite lengths, and we need not therefore take into 
 account the variation that would arise from the effect of tho 
 angle made with tho connecting rod. By the side of this we 
 see an engine cutting off at half-stroke, and you will see now 
 by Fig. 2 below it that the variation in tangential force is 
 very considerably increased ; Fig. 3 shows an engine cutting 
 off at one-eighth stroke, and you will see below the increase 
 in inequality of the tangential force. Fig. 4 shows a diagram 
 from a high and a low-pressure cylinder having the respective 
 cubic capacities of one and four, and wherein the steam is 
 cut off at one-half the stroke in the high-pressure cylinder, 
 so that it is expanded eight times, as in the case of the 
 single cylinder engine considered in Fig. 3. 
 
 Fig. 4 shows the aggregate effective pressures of the two 
 pistons at various points of the stroke, and below this the 
 variation in tangential results obtained from these cylinders. 
 There is no doubt whatever that by using two cylinders, 
 we have been able very much to reduce the irregularity 
 in the tangential force which existed with a single cylinder 
 when cutting off at one-eighth stroke. I will also call 
 your attention to another Diagram (11) showing the benefits 
 arising from having more cylinders than one with cranks 
 placed at different angles, so that with two cylinders when 
 one is on the dead point the other is doing its fullest duty. 
 An inspection of this Diagram 11 will show that with a pair 
 of cranks at right angles, and with two single cylinders 
 cutting off at half-stroke, there is a great approximation 
 to uniformity of tangential force throughout the whole revo- 
 lution. 
 
 Let us now consider whether it is worth while to go to the 
 complication of a double cylinder engine for the purpose of 
 obtaining this greater uniformity of tangential force. Nobody 
 objects to the use of a single cylinder non-expansive engine. 
 One is content with that, and no one uses the word irregu- 
 larity. I . should now like to ask you to bear with me for 
 a short time while I go into the question of the fly-wheel. 
 It is true that the fly-wheel cannot give absolute regularity 
 
 
 
160 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 of rotation ; in fact, if the result were absolute regularity 
 the fly-wheel would be an entirely useless implement. Its 
 object is to store up, when the rotation is in excess of the 
 mean speed, a certain amount of energy in order to deliver 
 it out when the rotation is below that mean speed : there- 
 fore, for it to be of any use, we must presuppose that a 
 certain amount of variation is to be permitted. What we 
 have to see is how much variation we can do with, and 
 
 DIAGRAM 11. 
 
 then see what kind of fly-wheel we ought to employ. Really 
 a, fly-wheel may be likened in its operation to a banker's 
 account. Supposing a man has an income of <! a day, 
 and he spends 1 a day, his income and expenditure 
 are equal and continuous, as he spends every day his income 
 as he gets it ; such a man does not want any balance at 
 his banker's at all. But suppose he spends 1 a day, and 
 gets his income paid to him once in ten days ; if he has not 
 
THE STEAM-ENGINE. 161 
 
 got a balance of .10 at his "banker's he is insolvent for 
 certain periods, and therefore he must have a 10 balance. 
 Supposing he gets his income paid every hundred days, he 
 must have 100 balance. Therefore, similarly as we get 
 more or less irregularity in the tangential force, so must we 
 have a larger or smaller fly-wheel, which we may look upon 
 as analogous to a banker's balance. I will take an instance, 
 and ask you to consider that we have a non-expansive engine 
 developing 240 gross indicated horse-power at 60 revolutions 
 a minute; that will be two horse-power for every half 
 revolution, equal in round numbers to 30 foot tons. Now 
 you will find if you calculate out that tangential diagram for 
 a non-expansive engine, that the variation above and below 
 the mean is about 30 per cent., and 30 per cent, of 30 foot 
 tons is 9 foot tons. This 9 foot tons, therefore, has to be 
 stored up in the fly-wheel during the time when we have the 
 excess of pressure, in order to be given out during the time 
 when we have the deficiency. Now if we have a fly-wheel 
 on this engine of such a size (about 13 feet diameter) that its 
 rim is going at 40 feet a second, and if we allow the weight of 
 the rim to be 9 tons, and assume the whole weight of the wheel 
 to be accumulated there, the value of that wheel will be 225 
 foot tons in round numbers. Now we want to store up 9 
 foot tons. It is a 9-ton wheel. We want, therefore, to pub 
 in that wheel a velocity equal to what it would derive in 
 falling from an extra foot in height. The velocity of 40 feet 
 a second is that which it would derive in falling from a height 
 of 25 feet, and we want to be able to put into it a velocity 
 which would equal that derived from falling from 26 feet, 
 which would be a velocity of 40*8 feet per second ; '8 is one- 
 fiftieth of forty, and therefore under the circumstances of 
 which I have spoken, we have a variation above and below 
 the mean speed of the engine of one-fiftieth of that mean 
 speed, a variation not inconsistent with very many purposes 
 to which engines are applied. If we were to use an 18-ton 
 wheel going at the same circumferential velocity, the variation 
 would be half of this fiftieth, and if we were * to use a 9-ton 
 wheel of double diameter we should get only one-fourth of 
 the variation, but if we were to use a fly-wheel of double 
 weight and double diameter we should get only one-eighth 
 of the fiftieth or one four-hundredth variation. 
 
 Keferring to the diagram of the expansive engine, cutting 
 
 H 
 
162 LECTURES TO SCIENCE TEACHERS. 
 
 off at one-eighth of the stroke, you will find that the varia- 
 tion of its tangential force is as much as 45 per cent, in excess 
 of the variation of the tangential force of the engine without 
 expansion. Now that excess of irregularity may be absolutely 
 neutralized by increasing the velocity of the supposed engine 
 from 60 to 69 revolutions. That is to say, supposing the 
 designer of the expansive engine wanted 240 gross horse- 
 power, he might either obtain it and a sufficient approxima- 
 tion to regularity by an engine running at 60 revolutions 
 without expansion, or he might so proportion the gearing 
 that it should run 69 revolutions. Were he so to pro- 
 portion it, it is obvious that the mean work done by the 
 piston at each revolution would be f ths of what he would 
 get in the first case, and taking this reduction of the mean 
 work and the increased effect of the fly-wheel, running at an 
 extra 9 revolutions, you will find that the same size and 
 weight of fly-wheel will cause this highly expansive engine 
 to work with the same approximation to regularity as that 
 with which the non-expansive engine running at 60 revolu- 
 tions would work. I think this fact is not sufficiently borne 
 in mind when engines are designed for driving any particular 
 kind of machinery. 
 
 There is no doubt that the double cylinder engine is the 
 engine of the present day. I do not think its economy really 
 lies in its principle, but that its economy in practice arises 
 from another thing altogether, and that is this, that by making 
 a double cylinder engine you put it out of the power of an 
 ignorant engine-driver to do away with that which you want 
 high expansion ; he must get high expansion ; he is com- 
 pelled to use it whether he likes it or not, whereas with the 
 single cylinder expansive engine he has the power to follow 
 the dictates of his own ignorance, and as a matter of obser- 
 vation I have hardly ever seen such an engine left to the 
 control of an engine-driver but it invariably worked at the 
 lowest possible grade, and, as I have said, I believe that this 
 withdrawal of control is to a large extent the secret of the 
 success of the compound cylinder engine. 
 
 Further with regard to expansion, I wish now to speak of 
 the question of the steam jacket. The steam jacket we owe 
 to Watt, and it was condemned by one of our earliest writers 
 on the steam-engine, Tredgold, who said it was merely adding 
 to the surface to be cooled by the air, but we know now that 
 
THE STEAM-ENGINE. 163 
 
 we cannot work with the expansive engines to any good 
 purpose without a steam jacket, and I will endeavour to show 
 you how that is. I have here (Diagram 12) a section of a 
 cylinder with a steam jacket round it, and this diagram will, 
 I trust, explain why the steam jacket is beneficial. Assume 
 the cylinder to be without a steam jacket, and that then you 
 
 DIAGRAM 12. 
 
 have high pressure and proportionately hot steam let into that 
 cylinder, the immediate effect is to condense a portion of the 
 steam and to make it into water. It disappears as steam, 
 and the indicator diagram will only inform us, we had received 
 the steam up to a certain point, whereas in truth another 
 portion has been received but, having been condensed, does 
 not appear on the diagram. As the steam expands and 
 
 M 2 
 
164 LECTURES TO SCIENCE TEACHERS. 
 
 becomes lower in temperature, the heat which has been 
 absorbed by the walls of the cylinder is given out, and the 
 dew which has been deposited upon them evaporates, and we 
 thus get a final pressure when there is no jacket, very much 
 in excess of that which ought to arise from the quantity of 
 steam that appears to have been put into the cylinder. You 
 will always find in a high expansive non-jacketed engine that 
 the final pressure is very much in excess of that which 
 apparently ought to be there, but if we take into account the 
 steam which has been condensed, you will readily see where 
 this excess of final pressure comes from, Now, had not this 
 portion been condensed, but remained as steam, instead of 
 obtaining only the useful effect represented by the full white 
 line on that diagram, we should have the useful effect repre- 
 sented by the middle dotted white line, which at its termi- 
 nation would coincide with the full white line, but in any 
 other position is outside of it. This difference shows the 
 excess of useful effect obtained by the aid of the steam 
 jacket. You also get the steam superheated by the steam in 
 the jacket during the progress of the piston. 
 
 I have spoken of the variations of tangential force, but 
 those are variations which occur while the resistance to the 
 engine is constant; we know that in factory engines, however, 
 the resistance varies very considerably, and therefore one is 
 compelled to have some kind of governor. I wish time 
 admitted of my going into the question, because we have 
 here an interesting series of governors. In the Watt governor 
 the number of revolutions at which the balls will fly out 
 depends upon the vertical height between the point where 
 the arms cross each other in the centre of the spindle and 
 the plane where the revolutions of the balls may be ; when 
 the load upon the engine is light, in order that the throttle 
 valve may be sufficiently nearly closed, the balls require to 
 be right up, and to maintain them there, the engine must 
 run at a pace in excess of that at which it runs when it is 
 heavily loaded. 
 
 From this cause the Watt governor, where the arms 
 vibrate on a pin in the axis of the vertical spindle, is not 
 such a true controller of the engine as to keep it at 
 regular speeds. The model by Mr. Jeremiah Head I have 
 here illustrates this very welL Here is another very common 
 form of governor which is extremely bad, where the pivot of 
 
 
THE STEAM-ENGINE. 165 
 
 the arm is on one side of the spindle; if we imagine the arms 
 produced until they intersect the axis of the spindle, we shall 
 see that when the balls go up, the point of intersection of the 
 arms with the spindle comes down, and in that way the vari- 
 ation in the vertical height of the governor is much greater 
 than when the joint of the arms is made in the line of the 
 spindle itself as in the Watt governor. But Mr. Head im- 
 proves upon the Watt governor ; he takes the arms through 
 the spindle and pivots them on its opposite side ; so that 
 when the balls rise, the point of intersection of the arms with 
 the spindle rises also, and that governor, therefore, will act 
 with comparatively little variation in its effective vertical 
 height. 
 
 Then we have Dr. Siemen's chronometric governor, which 
 acts as an absolute regulator, and we have also Dr. Siemen's 
 liquid governor, which also operates with absolute regularity. 
 I am only sorry it is impossible to enter into a description 
 of them. 
 
 Passing from this subject ; there are two things which in 
 these days everybody knows all about ; one is the wire-drawing 
 of steam, and the other is the freeboard of a ship. I am 
 not going to say anything about the latter, but in regard 
 to the wire-drawing of steam there is among the " every- 
 bodies " the strongest possible notion that it is a thing to be 
 deprecated. But far from it, in the ordinary construction of 
 bad engines, of which we have so many, wire-drawing is really 
 a thing to be admired. Consider the case of an engine 
 without any expansion valve, a cylinder far too large for 
 its work, with a governor having a throttle valve very 
 nearly closed, and with the steam going into the jacket of 
 the slide valve, and accumulating there until the valve opens 
 to let it into the cylinder. During the time the crank is 
 near its centres, and the piston is going slowly, then notwith- 
 standing the orifice for the inlet of the steam is restricted, an 
 accumulation of steam in the slide jacket is obtained so that 
 at that time the pressure there approximates to that in the 
 boiler ; when the slide valve opens that pressure comes upon 
 the piston, and thus we get a sort of irregular, very bad ex- 
 pansion diagram as the indicator figure ; but for what you do 
 get it is of value. On this point I must call your attention 
 to Diagram 1 3, in illustration of what may result from wire- 
 drawing in an engine demanding a mean pressure of 27 Ibs. 
 
IC>6 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 H 
 
 DIAGRAM 13. 
 
THE STEAM-ENGINE. 167 
 
 above atmosphere, without any expansion or condensation. 
 The quantity of steam employed, if there were no wire- 
 drawing, would be expressed by the area of Fig. I ; but 
 suppose now we wire-draw, and put a pressure of 37 Ibs. on 
 the piston at the beginning, then the line of pressure may 
 be represented by Fig. 2, and we have a reduced consump- 
 tion of steam as there shown. Supposing now we wire-draw 
 still further, and put a pressure of 57 Ibs. on the piston at the 
 beginning, we then reduce the consumption of steam to the 
 point shown on the third figure, and it really, therefore, 
 is a fact that where you have a bad engine you will do good 
 by wire-drawing. Do not understand me that it is a thing to 
 be used in lieu of proper expansion, but merely as saying 
 that so far from being, as is popularly supposed, a detriment, 
 in every instance, it is, in the case of a bad steam-engine, a 
 positive advantage. Again, take an ordinary engine without 
 expansion, and look at the enormous advantage which will 
 result from putting even a slight lap on the valve. Where 
 the slide travels about four times the amount of the steam 
 opening, you can. with a single slide, cut off at three-fourths of 
 the stroke, and by that arrangement you clearly introduce 
 into the cylinder only three-fourths of the steam, while you 
 have not diminished the work by even ^. Therefore, 
 in every bad engine, it is very easy for the first mechanical 
 engineer in the neighbourhood to do his part towards effecting 
 some saving in our store of coal. 
 
 Engines should be constantly examined by the aid of the 
 steam-engine indicator. The original indicator as devised by 
 "Watt is shown by the diagram on the wall, with the re- 
 ciprocating card travelling at right angles to the path of the 
 pencil caused to move by the pressure of the steam. This 
 indicator has been largely improved by Richards, as shown 
 in the further diagram. The improvement has consisted 
 in diminishing the piston stroke, and multiplying it by 
 means of levers, so that the momentum of the indicator 
 is greatly reduced. In that way, and without any alteration 
 of the indicator as ordinarily sold, it is possible to take 
 thoroughly good diagrams of engines running up to 300 revo- 
 lutions a minute. It fell to my lot, however, to have to take 
 a diagram from one of Thorneycroft's quick steam launches 
 running at 600 revolutions a minute, and I was very desirous, 
 for the purpose of a paper I was writing, of obtaining the vary- 
 
168 LECTURES TO SCIENCE TEACHERS. 
 
 ing power of these engines at from 200 up to 600 revolutions. 
 To do that it became necessary to take indicator diagrams, 
 and the way I set about it was by limiting the stroke of the 
 indicator, and by employing an extremely strong spring. I got 
 for 120 Ibs. steam only one inch rise on the figure. I then 
 limited the circumferential traverse of the indicator in the 
 same way ; T. provided it with an india-rubber spring, so as to 
 make it speak quickly, and in that manner, by having a 
 diagram not more than an inch in either direction, I obtained 
 most perfect and accurate figures from an engine running 
 at 600 revolutions. You will see them published in the 
 Transactions of the Institute of Naval Architects. By the 
 indicator we are able to tell, not merely the total work 
 that the engine is doing, but whether it is doing it economi- 
 cally and efficiently or not, and if not, are enabled to say 
 where an alteration ought to be made. The indicator gives 
 us the gross indicated horse-power, from which has to be 
 deducted the friction of the different parts of the engine. I 
 may say upon that, that at the Eoyal Agricultural Society, 
 when experimenting we had always used the dynamometrical 
 brake, the actual brake which is shown in this Exhibition. 
 On the last occasion, still using this brake, we put on indi- 
 cators, and were thereby enabled to obtain the ratio of effect 
 on the brake as compared with the indicated horse-power. 
 Speaking from recollection it averaged, taking one engine 
 with another, a difference of 17 per cent. These were non- 
 condensing engines, but I should call your attention to the 
 fact that they were worked in the excellent manner I spoke 
 of yesterday, and yielded results so high, that in one case 
 we got 79*49 million pounds raised one foot for 1 cwt. of 
 coal; a duty that probably is not attained by any single 
 engine in Cornwall, for I am sorry to say that the Cornish 
 engines have very much fallen off in their performances. 
 
 I should like now to say a few words about the injector. 
 We have here, cut open, one of the Giffard injectors so 
 commonly used for feeding boilers ; and I remember distinctly 
 when this instrument was first brought to my attention by 
 Mr. Robinson, of Sharp, Stewart, & Co., I and several en- 
 gineers were together, and, on being told by Mr. Robinson 
 that there was an implement by which the steam from a 
 boiler could be caused to generate a jet of water so powerful 
 as to enter that same boiler against the pressure within it, with 
 
THE STEAM-ENGINE. 16 9 
 
 the arrogance, I am sorry to say, common to many of us, 
 I said, " nonsense," and we all said " nonsense," and there 
 was a chorus of " nonsense," and nonsense it was, for we had 
 to go and see it at work, and there was no doubt about the 
 result. Then came the question why it worked. Why did 
 we see the steam re-enter the boiler from which it had 
 started 1 There have been a great many papers written upon 
 this subject in a mathematical form, but I have said on 
 more than one occasion, that I arrogate to myself, rightly or 
 wrongly, the credit of being the first to give a popular ex- 
 planation of the action of the injector. I believe the whole 
 theory is to be explained by one word, and that is " con- 
 centration." Let me illustrate what I mean. Supposing 
 there were a cistern containing water with a nozzle upon the 
 side of it near the bottom, from which the water could issue 
 horizontally in a jet. Assume the jet were opposite to an 
 orifice in an empty cistern parallel with the first. We 
 know that the jet of water would enter that orifice, that 
 the water would accumulate in the cistern which the jet 
 entered, and that the jet would continue to enter until 
 the accumulated water in the second cistern caused the jet 
 to have such an opposition that it could not enter. Now 
 Mr. Froude has shown us in his experiments that when you 
 have once got the water into motion, if you will allow a 
 sufficient difference of head between the two cisterns to 
 compensate for the resistance caused by the friction in the 
 tubes through which the water passes, the water will, if you 
 have a proper nozzle, continue to enter into the second cistern 
 when it is full up to the height of the first one, minus the 
 difference expressing the friction. We see, therefore, that we 
 may get the water entering the second vessel very nearly to 
 the height from which it issued from the first. But supposing 
 that the water in its passage across the imaginary space from 
 the vessel from which it flowed, to the one into which it is to 
 enter, could by some process of magic be converted into 
 mercury, the same weight of mercury as was equal to the 
 weight of water : we know that if we could do that we 
 should diminish the area of the stream to yj- or thereabouts, 
 so that the stream which came out of an orifice of a square 
 inch, when converted into mercury would enter the other 
 tank through an orifice of ^ of a square inch ; but the 
 stream being of the same weight and flowing with the same 
 
170 LECTURES TO SCIENCE TEACHERS. 
 
 velocity, would obviously concentrate on that -^ square inch 
 the whole of the energy which, if it remained water, would 
 otherwise have been applied on the square inch, and thus 
 by this act of concentration there would be 14 times the 
 power per unit of surface, and it would come with the same 
 effect as if it came from 14 times the height. With such a 
 concentration it would enter into the second tank not merely 
 against a head equal to that from which it came, but against 
 a head very nearly 14 times as great; and you see, therefore, 
 that if you could by magic convert water into mercury, you 
 would then get the power to enter against a very excessive 
 pressure to re-enter its own tank, and much more. Now, 
 instead of taking water issuing from a tank, let us take steam 
 issuing from a boiler, and that steam without magic we can 
 concentrate, because we can condense it, and having con- 
 densed it, we do not concentrate it only 14 times, but 1,696, 
 or 388, or 144 times, or whatever may be the ratio of the 
 particular steam to water. By that concentration, therefore, 
 we have an enormous power, competent not only by itself 
 to re-enter the boiler from which the steam came, but a 
 power so much in excess of this as to be able to take along 
 with it the water employed in producing the concentration, 
 that is the condensation of the steam. In that way the Giffarcl 
 injector does introduce into the boiler the steam that came 
 out of it, and the water that condensed that steam. 
 
 Except in one solitary instance (that of the surface of 
 boilers) I have abstained from giving any proportions ; still 
 less have I had opportunity to indicate the relations between 
 dimensions and horse-power, and I may say that had I had 
 time to dwell upon this branch of the subject of the steam- 
 engine I should have been met with the curious difficulty of 
 not being able to say what in manufacture is meant by a 
 horse-power. But I may tell you now, that, among the five 
 kinds of horse-power which are recognised (for there are five), 
 I, whenever I have used the term horse-power, have referred 
 to the engineer's horse-power of 33,000 Ibs. (roundly, 15 
 tons) raised through one foot in a minute of time. 
 
 Dr. Joule and others have shown us that 772 foot Ibs. 
 represent the equivalent of one heat unit, and we know, 
 there fore,t hat 33,000 foot Ibs. per minute ought to be got for 
 an hourly expenditure of 2,565 heat units. Llangennech coal 
 is capable of giving forth 14,718 heat units per lb., and 
 
THE STEAM-ENGINE. 171 
 
 we ought therefore by an expenditure of '17 Ibs. of that coal 
 to obtain one horse-power for an hour, while the very lowest 
 consumption that I know of is 1J lb., therefore the ratio be- 
 tween those two figures is 9 to 1, so that at the present time 
 the best engines use but one-ninth of all the heat which is 
 theoretically given forth by the coal. We see, therefore, 
 there is before mechanical engineers a very large field for 
 improvement. Do not imagine that I am suggesting that we 
 shall ever obtain the whole of the theoretical effect or any- 
 thing bordering upon it. Certainly not by the steam-engine, 
 for we do not yet know of materials which would admit of our 
 commencing with a pressure where the steam would be of a 
 density equal to that of water, nor could we tolerate cylin- 
 ders of such size as to enable the expansion to be carried on 
 until the temperature of ice was attained ; and yet those two 
 conditions, I believe, are necessary before we can even theo- 
 retically get as a useful effect the whole heat of the fuel, but 
 the efforts of mechanical engineers and of men of science for 
 the improvement of the steam-engine have not been barren of 
 results in the past, and will not, I trust and believe, be fruit- 
 less in the future. We are led to be hopeful when we recol 
 lect that within the last fifteen years the consumption in 
 ocean steamers has been brought down from 5 Ibs. of coal per 
 gross indicated horse-power, to 2J, 2, and even 1J Ibs. in 
 real regular ocean steaming; and we have in the case of her 
 Majesty's ship Briton when steaming at half-power, seen the 
 consumption as low as 1 J Ibs. of coal per gross indicated horse- 
 power per hour, which is, as I have said, the lowest that has 
 come under my notice in a real working engine. Is it too 
 much to hope that greater economy will yet be reached ? 
 
 I must now bring these two lectures to a close, and in 
 order that they may lead to some practical result I will at 
 this, their very conclusion, ask you, bearing in mind the 
 progress which has already been made, to look at the pair of 
 indicator diagrams to which I now point. These show the 
 immense difference in consumption of steam (and therefore of 
 coal) between two engines, the one (of no progress) working 
 non-expansively and non-condensing, and the other (the 
 engine of progress) working expansively and condensing. 
 Too frequently, from a desire both in the engine-maker and 
 in the engine-user to save a few pounds in the first cost, 
 wretched machines such as the non-expansive non-con- 
 
172 LECTURES TO SCIENCE TEACHERS. 
 
 densing engine from which this diagram was taken, are scat- 
 tered over the country and are wasting the fuel which it is 
 our duty to preserve for those who will come after us. If 
 you will bear these facts in mind and consider the loss indi- 
 cated by the diagram of the " engine of no progress," as I 
 have termed it, I shall feel that my labours here have pro- 
 duced some practical results. 
 
 I cannot hope that within the limits of these two lectures 
 it has been in my power to bring before you new matter, 
 nor, indeed, is it probable that anything I could say would 
 be absolutely new to an audience such as that which I have 
 just been addressing; the utmost I can hope, therefore, is 
 that these lectures may have been suggestive, and that by 
 bringing to your minds certain points in the construction 
 and management of the steam-engine, and by impressing you 
 with the fact that upon the attention paid to such points 
 economical results depend, I may be the means of inciting 
 each one of you when returned to his own district to exert him- 
 self to improve the steam-engine standard in that district, and 
 thus to get rid of the wasteful machines to which I have 
 just alluded ; and let me remind you that by so doing each 
 of you may perform his part of the duty that devolves upon 
 scientific mechanical men, viz., that of preventing the scan- 
 dalous waste of fuel that now, alas, too frequently occurs. 
 
 I must apologize to you for the audacity exhibited in my 
 endeavouring to lecture on a subject so extensive as that of 
 the " steam-engine " in a course of only two lectures, but let 
 me plead in extenuation that the audacity is not mine but 
 that of the Committee of Council on Education. 
 
 In conclusion let me thank you for the forbearance you 
 have shown, and also for the very great attention you have 
 paid to me. 
 
RADIATION. 
 
 BY PROFESSOR G. FORBES. 
 
 AMONG the different physical researches which are illustrated 
 by the collection of instruments at present exhibited, theie is 
 one which has performed a very important part in forward- 
 ing the progress of a special branch of physics. 
 
 This department, which is shown pretty fully by the 
 apparatus on the table, is the department of radiant heat, 
 y esterday we were discussing the question of the undulatory 
 theory to a certain extent, and we saw that by the measure- 
 ment of the velocity of light it was absolutely proved that 
 Newton's corpuscular theory of light was untenable, and 
 that therefore, as far as that theory was opposed to the 
 undulatory theory, the undulatory theory must be accepted 
 as true. There are a vast number of other facts, however, 
 which tend to prove the truth of the undulatory theory, both 
 directly and indirectly. Yet there is no proof so absolutely 
 certain of this theory as others, of the theory of gravitation 
 for example. In fact, in physics there are wonderfully few 
 theories which are absolutely demonstrated with the same 
 certainty as the law of gravitation. The law of gravitation 
 as stated by Newton is simply a statement of the facts 
 which have been observed, assuming the definition of force 
 as he gave it. It is not a theory it is simply a statement 
 of observation, a gathering together of a host of phenomena 
 into one general statement. Just as the whole of the pro- 
 positions included in spherical astronomy and spherical 
 trigonometry are dependent on the simple statement that a 
 sphere is a surface, every part of which is equally distant 
 
174 LECTURES TO SCIENCE TEA GHEES. 
 
 from the centre ; so also all the phenomena observed and 
 grouped by Newton are included in the statement of the 
 law of force as he denned force to be. 
 
 But there is no other theory in physics so firmly 
 demonstrated as this. Our belief in the truth of the 
 urdulatory theory rests on the fact that it explains a 
 vast number of totally different phenomena in a perfectly 
 clear way ; in fact, there are no phenomena in the theory of 
 light which, if they have not been explained by the 
 undulatory theory, we are not justified in saying could be 
 explained by it if our mathematical analysis were suf- 
 ficiently powerful to translate the meaning of the theory 
 completely. But the undulatory theory of light has 
 acquired certainty from a vast number of independent 
 sources as well. In order that the radiations from a 
 luminous body may be propagated through the ether, it is 
 necessary that the luminous body must have its molecules 
 in a state of rapid vibration, and not only has the whole 
 science of spectrum analysis led up to this, to point out 
 that the molecules of different bodies are vibrating in 
 certain definite periods, such as will give vibrations off to 
 the ether, and ultimately send them so as to strike the eye, 
 but also a number of theories which have been reached 
 independently, all tend and converge to this same point, 
 and show us that not only are the molecules of a body in 
 a state of rapid vibration, but that there is an ether 
 capable of transmitting those vibrations. The splendid 
 experimental results obtained by Professor Graham in 
 connection with the diffusion of gases have proved almost 
 beyond a doubt that the molecules of gases are moved about 
 with a very great velocity, and means have been devised 
 actually to measure the very velocity with which they are 
 moving. The researches of Clausius, Clerk Maxwell, 
 Thomson, Eankine, and others, have reduced the science of 
 the molecular motion of gases to an absolute certainty, so 
 far as theory can be certain, and they have explained tho 
 laws of gases perfectly upon this assumption. The very 
 size of one of these molecules* has been determined by Sir 
 William Thomson in four different manners, all of which 
 agree, within fair limits, with each other ; and from 
 arguments which it is impossible for me to go into at 
 present, Sir William Thomson has shown what is in all pro- 
 
EADIATION. 175 
 
 bability the actual nature, form, and motion of those 
 molecules. He shows that an atom is in all probability 
 nothing more nor less than a vortex ring. You have all 
 seen smoke rings which have been formed from the mouth 
 by tobacco smoke, or which are often seen coming from the 
 mouth of a cannon a ring of smoke passing through the 
 air with considerable velocity, and this ring having con- 
 volutions from the exterior of the ring into the interior 
 round and round. Such a smoke ring is the fundamental 
 idea of a vortex atom as denned by Sir William Thomson. 
 These results may seem to be too hypothetical, but on a 
 clear examination of the grounds which have led him to 
 adopt these views, there is very little doubt that if not the 
 true explanation of the nature of a molecule it is something 
 very close to it. 
 
 Furthermore as to the existence of the ether, we have 
 several independent proofs. The illustrious astronomer 
 Encke found that the comet which bears his name was 
 retarded year by year. 
 
 Each time this comet appeared in its elliptic orbit round 
 the sun, it was found to have a shorter period, and to be 
 revolving quicker round the sun. Now we know from the 
 law of gravity, that if it be revolving quicker round the 
 sim, it must be closer to the sun, and the only force which 
 Encke could conceive of to bring the comet close to the 
 sun was the resistance of an ether to its motion which 
 increases the force which the sun exercises upon it in 
 relation to its momentum forwards. Various astronomers 
 have calculated the effects of this, and the general 
 impression is that there is a resisting medium acting upon 
 this comet, but that the exact law of resistance which was 
 assumed by Encke is perhaps not exactly correct. 
 
 But we have furthermore some most remarkable experi- 
 ments by Professors Tait and Balfour Stewart, which appear 
 to point most clearly to the existence of an ether. Here 
 we have the remarkable apparatus which was employedby 
 them. It consists of a receiver which can be exhausted by 
 means of a powerful air-pump. Within this receiver there 
 is a disc which is capable of rapid rotation, and mechanism 
 is applied to it by means of which this rotation can be 
 given. A rod acts upon this mechanism and passes down 
 through a barometer tube, so that it can be easily turned 
 
176 LECTURES TO SCIENCE TEACHERS. 
 
 by hand or by other mechanism, and force can be applied to 
 the disc without affecting the vacuum at all. There is also 
 an instrument which I will describe presently, called the 
 thermopile, and by an ingenious arrangement of a baro- 
 meter-tube a motion can be given to the thermopile without 
 affecting the vacuum These gentlemen experimenting with 
 this apparatus exhausted the receiver of air* as completely 
 as it was possible to do, and then put the disc into very 
 rapid rotation for a considerable time, and testing it with a 
 thermopile, it was found to be heated. I cannot go into 
 the precautions which they took to make sure that this 
 heat was not due to friction or to the resistance of the air, 
 and so forth ; but they took careful account of these things, 
 measured the amount of these sources of heat as far as 
 possible, and allowed for it, and still there was outstanding 
 an amount of heat which could not be accounted for except 
 on the assumption that it was produced by the friction of 
 the ether. 
 
 Now the series of researches which we illustrate to-day 
 is one which connects the undulatory theory of light with 
 radiations generally, and with heat radiations in particular. 
 The first experiments which proved that there was some 
 analogy between heat and light were probably made by 
 Porta in the sixteenth century. He employed mirrors to 
 reflect heat, arranged as you see them here. He was able 
 to converge the heat to a focus, just as you can converge 
 light to a focus, and this experiment was repeated by Pictet 
 at the beginning of the present century. Pictet employed 
 the very mirrors which you see before you. He so arranged 
 them that the light placed in the focus of one mirror was 
 seen to throw an image of itself by the double reflection 
 from these two mirrors upon a certain definite point. 
 Having found that these were in the luminous focus, all that 
 remained was to show that they were in the heat focus 
 also ; and in order to do this Pictet employed a freezing 
 mixture, and showed not only that the heat was reflected, 
 but apparently that the cold was reflected too. If we 
 place a freezing mixture in the focus of one mirror we shall 
 probably be able to see that the air thermometer in the 
 focus of the other shows a sensible production of cold. But 
 that does not mean really that cold is radiated from the 
 freezing mixture, but that heat is being radiated from the 
 
 
RADIATION. 177 
 
 bulb. Heat is radiated from every part of the room, from 
 the freezing mixture as well as from the walls, but there is 
 less heat radiated from the freezing mixture than from 
 other parts, whereas the bulb of the thermometer will be 
 radiating heat not only to the walls of the room, but also 
 to the mirror and thence to the freezing mixture. There is 
 a continual exchange of heat going on between the thermo 
 meter and surrounding bodies, if they are all at the same 
 temperature the thermometer does not change, because it 
 
 FIG. 1. 
 
 is radiating as much heat as it is absorbing; but if we 
 place a freezing mixture in that focus, the thermometer 
 will radiate to that freezing mixture a certain amount of 
 heat greater than the amount of heat it gets from it. 
 
 After this was stated by Pictet in Geneva, there being a 
 large band of philosophers bound together at Geneva, they 
 continued to take up this, and to examine the laws of 
 radiant heat with some care. The most important results 
 that were obtained were those obtained by M. Prevost. 
 He published a theory of the law of exchanges which has 
 
 N 
 
178 LECTURES TO SCIENCE TEACHERS. 
 
 been amplified by others and which was brought before the 
 world in the writings of Professor Balfour Stewart. 
 
 The theory of exchanges is simply that which I was 
 speaking of just now that we are not to consider when the 
 thermometer, say, is heated by radiation, that it is simply 
 receiving radiations from a hot body, but also that it is 
 giving off radiations at the same time, that it is radiating 
 its heat to surrounding bodies. When the thermometer 
 does not move, we know that there is an equilibrium of 
 temperature and that consequently the quantity of heat it 
 is radiating is equal to that which it is receiving. If it 
 rises we know that it is receiving a greater number of radia- 
 tions than it is giving off ; and on the contrary, when it is 
 falling it is radiating more than it is absorbing. The theory 
 which Prevost enunciated, called the theory of exchanges, 
 leads us directly to several very important results which 
 were proved to follow from this theory partly by Prevost 
 and partly by Fourier. These facts are stated in the follow- 
 ing table : 
 
 1 st. All surfaces absorb as much heat as they radiate at 
 the same temperature. 
 
 2nd. The quantity of heat radiated increases with tem- 
 perature. 
 
 3rd. The heat radiated varies inversely with the square 
 of the distance. 
 
 4th. It also varies as the cosine of the angle of radiation. 
 
 The most important of these propositions is the third 
 which was proved by Fourier ; but we will take them in order. 
 I have already pointed out the reason of the first one ; because 
 if this bulb did not radiate the same quantity as heat as it 
 absorbs at its present temperature suppose it radiated more 
 heat than it absorbed then although all objects around it are 
 at the same temperature, still we should see the thermometer 
 gradually falling ; whereas if it radiates less heat than it 
 absorbs, the bodies all round it being at the same tempera- 
 ture, we should find that the thermometer would indicate an 
 increase of temperature and it would be gradually rising. 
 The second law, that the quantity of heat radiated increases 
 with the temperature is evident, because the surrounding 
 bodies are hotter than this thermometer. The thermometer 
 increases in temperature, therefore the thermometer is 
 absorbing more heat than it is radiating; therefore the 
 
RADIATION. 179 
 
 surrounding bodies are radiating more heat to it than they 
 would radiate to it if they were at the same temperature as 
 the thermometer itself. Therefore we know that these 
 surrounding bodies, when they are at a greater temperature 
 are radiating more heat than they would if they were at a 
 lower temperature. The third law was shown by Fourier also 
 to follow from the theory of exchanges at equilibrium of 
 temperature. He supposed a large heated surface to shine 
 upon a thermometer or upon a thermopile. Suppose that 
 wall were heated, this instrument receives, when close to it, 
 a certain amount of intense heat from a small surface of the 
 wall. "When I double the distance the surface of the 
 wall which is radiating into the thermopile is four times as 
 great, and when I treble the distance, is nine times as great. 
 The surface of the wall radiating heat into the thermopile 
 will always be proportionate to the square of the distance ; 
 but upon making the measurements experimentally we find 
 that the temperature registered by the thermopile is the same 
 in all cases. In consequence of this, since the surface which 
 is radiating heat increases with the square of the distance, 
 we know that the intensity of radiation from any portion of 
 that surface must diminish in proportion to the square of 
 the distance. The fourth law is also proved in the same 
 manner, because if I keep the thermopile constantly at 
 the same distance and incline it to the wall giving off 
 the heat, we always find the temperature registered the 
 same. Now when it is perpendicular to the wall there is a 
 small surface of the wall giving off heat ; but when I 
 incline the thermopile at an angle there is an elongated 
 space in the wall giving off heat, and yet the temperature 
 is the same. The surface is diminished in this case in pro- 
 portion to the cosine of the angle between these two 
 directions, and since the temperature is the same, although 
 the surface is lessened in the ratio of the cosine of the 
 angle, it follows that the intensity of radiation from any 
 portion of the surface is proportional to the cosine of that 
 angle from the perpendicular. 
 
 Some of the most important experimental results in this 
 subject were those obtained by Professor Leslie, of Edin- 
 burgh. Unfortunately we have none of the actual apparatus 
 which he employed; but the instrument which he made 
 use of was a curve of this sort, which was called a differ- 
 
ISO 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 ential air-thermometer. There is a bent tube with liquid 
 in the centre of it terminating in two bulbs. If one of 
 those bulbs is heated the air in it expands considerably, 
 and this will be shown by a rising of the liquid in one arm 
 of the tube and a sinking in the other arm, whereas if it 
 were cooled the contrary would take place and the liquid 
 would actually descend in this arm and rise in that arm. 
 This instrument, together with an apparatus called a 
 Leslie's cube, which consists of a cube of metal with sur- 
 
 Fio. 2. 
 
 faces of different natures, some highly polished, others 
 blackened or whitened, and so forth, were the chief things 
 employed by Leslie. He arrives at the result that different 
 surfaces radiate heat to different extents. Thus if he em- 
 ployed a polished surface and heated the cube with boiling 
 water, the thermometer was much less heated by radiation 
 than if he employed either a, black surface or a white sur- 
 face \ in fact, he drew out a table of a large number of 
 different substances, and showed the relative different 
 
RADIATION. 181 
 
 radiations from these substances when a source of heat 
 which was not luminous was employed. 
 
 These researches led Dr. Wells, in the year 1815, to 
 his theory of dew. Dr. "Wells observed that if radiations 
 are taking place from all surfaces of the earth from the 
 surface of grass, metals, and other objects then when the 
 sun is away so that the atmosphere is not heated, those 
 objects ought all to get cool by radiation, throwing heat oft' 
 into space. But following out the conclusions of Leslie, he 
 said there are some substances which ought to radiate 
 heat very much more than other substances. For instance, 
 grass and the leaves of trees, or the substance of which a 
 spider's web is composed, are all good radiators, and those 
 ought to get cool quicker than a metal or any substance 
 like that which does not radiate so much. He 'also saw 
 that if any object be covered at a little height, however 
 thin that covering be, it will prevent free radiation from 
 the surface of the ground to the sky, and consequently it 
 will not cool so rapidly. He noticed that those were all 
 the conditions which were favourable to the deposition of 
 dew, and he saw in this an explanation of the formation 
 of dew. The atmosphere always contains a certain amount 
 of moisture ; if it were to be suddenly cooled the amount 
 of moisture which it would be capable of -containing in the 
 form of vapour would be sensibly diminished. If now in 
 the atmosphere there be a sufficient amount of moisture so 
 as to saturate it, and suppose that to be cooled, it will no 
 longer be able to contain this moisture in the form of 
 vapour, and consequently it will be deposited in the form 
 of dew on any surface which has cooled down to below 
 the dew-point of the air at that time. I shall be able to 
 make this clearer perhaps with the aid of a very rough 
 experiment. 
 
 You know that if one breathes upon a piece of glass it 
 will be covered with vapour deposited in the form of 
 minute drops of water, the reason being that the breath is 
 saturated with aqueous vapour. The glass when the breath 
 falls upon it is cold ; at that temperature the air can no 
 longer suspend so much moisture in the form of vapour, 
 and consequently it must be condensed. This, then, illus- 
 trates the manner in which dew is formed on surfaces 
 
182 LECTURES TO SCIENCE TEACHERS. 
 
 which radiate most heat ; but if, previously to breathing 
 upon the glass, we slowly heat it so as to raise it to the 
 temperature of the body, then when we breathe upon 
 it we do not cool the breath below the temperature of 
 saturation, and we shall find that it can no longer 
 condense the vapour which is contained in the breath. 
 The glass is absolutely clear and transparent after being 
 breathed upon. 
 
 Simultaneously with these advances in the science of 
 radiant heat, some very curious researches were being made 
 which tended to divide heat into different kinds. Sir 
 William Herschel had observed that the temperature of the 
 spectrum varies in different parts. He formed a spectrum 
 with a prism of glass, and, passing a thermometer along 
 the different parts of the spectrum, he found the tem- 
 perature increase as he approached the red end, and not 
 only so, but he got a very large amount of heat in the 
 part which was utterly invisible beyond the red. This 
 was a very important advance, and was taken up after- 
 wards by other celebrated experimenters by Seebeck, the 
 illustrious discoverer of thermo-electricity, who experi- 
 mented on light and other subjects. He found that the 
 position of the maximum point depended on the nature of 
 the prism which was employed ; thus when he employed a 
 prism of flint glass he found the maximum was beyond the 
 red, but when he employed one of water he found that the 
 maximum was in the yellow. The next person who threw 
 light on this subject was Delaroche. He employed a 
 luminous source of heat, and measured the effect which it 
 had upon the thermometer, but he found that as soon as 
 he placed a piece of glass between the luminous source of 
 heat and the thermometer, the thermometer indicated an 
 increase of temperature veiy much less ; that there was 
 only three or f our per cent, of heat transmitted through 
 the glass ; but when he employed heat which had already 
 passed through one piece of glass, then on putting another 
 piece of glass in front of it, he found that the temperature 
 was hardly diminished at all ; he found that heat, in fact, 
 which had once passed through a piece of glass could easily 
 pass through another piece of glass also. Thus there seemed 
 to be some reason for dividing heat into two kinds one 
 
RADIATION. 
 
 183 
 
 which could pass through glass and the other which could 
 only pass with difficulty : the effect of introducing the first 
 piece of glass was to cut off one kind of heat. This subject 
 was fully investigated in a splendid series of researches by 
 Melloni ; he was led to take up these researches by asso- 
 ciating himself with Nobili, an Italian, who has the merit 
 of inventing this beautiful instrument called the thermo- 
 pile, an instrument which renders experiments in radiant 
 
 FIG. 3. 
 
 heat practicable in many cases where they would not be so 
 without it. I will not waste time in describing the prin- 
 ciple of the thermopile, except to say briefly that there are 
 a number of bars of antimony and bismuth joined together 
 in pairs alternating at the two ends. It had been found 
 
184 L EGTUEES TO SCIENCE TEA CHEES. 
 
 by Seebeck that if either end were heated more than the 
 other, a current would pass through this continuous series 
 of bars and through a wire which might be connected with 
 the extremity of the series and with a galvanometer, so 
 that all that was necessary in order to measure the increase 
 of temperature in one side over the other was to measure 
 the intensity of the current of electricity passing through 
 the wire. Here is another thermopile, and we shall see 
 presently, as soon as the beam falls on the surface, a 
 current of electricity will pass through it and be indicated 
 by the motion of the galvanometer. Employing this instru- 
 ment, Melloni was led to other important results. The 
 most important perhaps was his grand discovery that rock- 
 salt is a substance which is apparently absolutely perme- 
 able to heat. He believed there was no loss of heat by 
 absorption when radiant heat was passing through rock- 
 salt, but it is better to say that there was no absorption 
 perceptible with the instrument that he employed. Un- 
 fortunately rock-salt is a substance that deteriorates very 
 much by exposure to a moist atmosphere, and consequently 
 it cannot be employed in laboratories except under peculiar 
 circumstances, and it has to be preserved in a very dry 
 atmosphere. I have here a lens, a prism, and a plate of 
 rock-salt, which are exhibited in the Loan Exhibition by 
 Stegg, of Hamburg ; and in order to preserve them they 
 are surrounded by pieces of glass. Almost all the great 
 researches in radiant heat have been made by the aid of 
 rock-salt apparatus. Melloni measured the amount of heat 
 which is found in different parts of the spectrum, and also 
 contained the method which was employed by Delaroche 
 of sifting heat. "We find his results included in a consider- 
 able-sized volume called La Thermochrose, or Heat Colora- 
 tion, we may call it ; and he used flames and sources of 
 heat of different kinds. He found different results from 
 different kinds of heat. Thus when he employed a flame 
 sent through rock-salt he found only 92 per cent, got 
 through, some being reflected from the surface. "When he 
 employed plate-glass only 39 per cent, got through, and 
 with alum only 9 per cent. Then when he employed other 
 sources of light he got the heat proportionately, as shown 
 in this table : 
 
RADIATION. 
 
 MELLONI'S EXPKRIMENT. 
 
 185 
 
 
 Flame. 
 
 Incandescent Surface at 
 Platinum. | 700 F. 
 
 Heat at 
 212 F. 
 
 Eock Salt. . 
 
 92 
 
 92 
 
 92 
 
 92 
 
 Plate Glass . 
 
 39 
 
 24 
 
 6 
 
 
 
 Alum . . . 
 
 9 
 
 2 
 
 
 
 
 
 You will observe that with non-luminous heat, alum 
 refuses to allow any heat to pass through it at all, although 
 it is a substance which is perfectly transparent. Plate- 
 glass is a substance which lies half-way between rock-salt 
 and alum, the loss being only about 10 per cent. ; so that 
 he got 90 per cent, of the heat through the second piece of 
 alum, clearly showing that there are certain rays which 
 are absorbed by the alum, and all the other rays are allowed 
 to pass quite freely through it. In fact, he saw that every 
 substance had the property of absorbing certain colours of 
 the spectrum, and that heat differed in no degree from 
 light ; that there was a heat spectrum as well as a light 
 spectrum, but that the great intensity of heat was as a 
 general rule in parts of the spectrum which were invisible ; 
 and the reason of this is simply that the substance of which 
 the eye is composed does not allow these radiations of heat 
 to pass through. The general results of Melloni's work 
 may be represented by a diagram something of this sort in 
 which the upper part shows the curves representing the 
 intensity of radiation at different temperatures with the 
 curve of limiting visibility, whilst in the diagram beneath 
 is given the resulting curve of apparent visibility. The 
 thing wanting to show the absolute identity of light and 
 heat was some experiment on the phenomena of heat 
 analogous to those which had been performed on the 
 theory of light which acquired the name of polarised. 
 You know that if we employ several pieces of this sub- 
 stance called tourmaline, which is tolerably transparent, 
 if we first pass the light of a candle through one piece and 
 then through another, we can turn this piece of tourmaline 
 about in a certain direction until we get total darkness ; 
 that is to say, when the two pieces of tourmaline are 
 
180 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 arranged in a certain position no light can pass through 
 them, but if you turn one of them at a right angle then 
 the light can pass freely through both of them, and that is 
 the phenomena called polarisation. It was completely 
 explained by the undulatory theory of light and by that 
 theory alone, and it became desirable to see whether such 
 a thing could be done with heat from a low source of tem- 
 perature which was not luminous. Various experiments 
 had been made in the early part of the century to test this, 
 but with only negative results ; until hearing of the em- 
 ployment of this new form of instru- 
 ment for measuring temperature, viz. 
 the thermopile, the late Principal 
 Forbes employed the instrument in a 
 re examination of this question. He 
 had already tried the experiment 
 with other kinds of thermometers 
 with negative results, but on employ- 
 ing this thermopile with two pieces 
 of tourmaline he was enabled to show 
 the fact that there was polarisation, 
 and this having been once established 
 seemed to make it at once certain 
 that light and heat were absolutely 
 identical, because the source of heat 
 which he employed was a vessel of 
 boiling water which was non-lumin- 
 ous; but he continued to examine 
 the question further, and at last ob- 
 tained the means of measuring with 
 very great accuracy the amount of 
 polarisation. Here are two of the 
 instruments which were actually em- 
 ployed in these researches. They consist simply of a 
 cylinder and a piece of mica inclined at a proper angle 
 inside it. You know that if we take a large number of 
 plates of glass and lay them one over the other we can 
 reflect light at a cei*tain angle or transmit it at the same 
 angle, and the portion of light transmitted or reflected is 
 polarised ; and the larger the number of plates of glass, 
 the more complete is the polarisation. In order to get an 
 analogous phenomenon in the case of heat, he employed 
 
 Flu. 4. 
 
E AVIATION. 187 
 
 mioa because he had observed that mica transmitted a 
 large proportion of non-humous rays. He devised means of 
 splitting mica into a large number of minute plates by 
 exposing it to the action of heat. By so doing he was 
 able to keep the mica in one mass, and yet to have that 
 split into a large number of plates, and this was tho 
 material employed which led to successful observations 
 being made on the polarisation of heat. Before illustrating 
 this I will point out the table of Principal Forbes' results, 
 which is as follows : 
 
 FORBES' EXPERIMENT. 
 Refractive Index of Rock Salt. 
 
 Heat from lime, mean value. . . .1*531 
 Heat passed through glass .... T547 
 Heat passed through alum . . . . 1 '558 
 Mean luminous ray 1 '562 
 
 This proves in another way the identity of luminous and 
 non-luminous vibrations. By employing the angle of total 
 reflection with a prism of rock-salt, and measuring this 
 angle, Principal Forbes was able to find by simple geome- 
 trical principles the refractive index of rock-salt for dif- 
 ferent sources of heat. He first used the heat of lime, 
 then, after passing it through glass, he found a different 
 refractive index; and then when he passed it through 
 alum he got a third value, and from the mean luminous 
 ray he got a fourth value. You see the only difference is 
 that the lower the source of heat the more near it is to the 
 red end of the spectrum, or beyond the red end the greater 
 is the refractive index, which is in accordance with theory. 
 
 I will now try to show you with this apparatus the 
 experiment on the polarisation of heat. Unfortunately 
 these instruments, which were employed by Principal Forbes, 
 and which have been lent by Professor Tait, are in an 
 extremely fragile condition, and are also covered with dust. 
 I hardly like to run the risk of injuring them by wiping 
 the dust off, and therefore I will use a slightly different 
 form of apparatus, which will answer the same purpose. I 
 have here two large prisms which have been lent by Prof. 
 Guthrie. By means of this large Nicol prism we can 
 get a polarised beam of light. This is called the polariscr, 
 
188 LECTUBES TO SCIENCE TEACHERS. 
 
 and the next is called the analyser \ and when the light is 
 passed through this it is in a peculiar condition it is said 
 to be polarised ; but if after that I turn this round its axis 
 in different directions, I shall find that in some positions it 
 will cut off all the light, but in others it will allow a con- 
 siderable amount of light to pass. If our adjustments 
 were perfect, it would now have totally disappeared, but 
 there is a difficulty in preventing a slight reflection from 
 the interior of this prism. Now I wish to show the same 
 phenomena with respect to heat ; but before doing so I 
 shall show you another experiment which we also have 
 with regard to light. When we have turned this round 
 until we get a total obscuration of light, or very nearly so, 
 I will pass between the polariser and the analyser a piece 
 of crystal mica, and that has the effect of altering the 
 condition of this ray and depolarising it so that it is again 
 at liberty to pass through this second prism. You see the 
 colours there, showing that the light is again passing 
 through, and by rotating the mica I can get different 
 colours. In the case of heat it will be most important 
 to employ those colours which are of a deep red, and then 
 we shall be more likely to get non-luminous rays trans- 
 mitted. You see first the light is polarised by this prism, 
 then on examining it with the second prism we find that 
 in certain positions it is allowed to pass by transmission, 
 whereas in others it is stopped out entirely ; but when it is 
 stopped out we get the light passing through the piece of 
 mica again Principal Forbes employed a non-luminous 
 source of heat, viz., boiling water ; but since we are un- 
 able to employ the same delicate means which he had, on 
 account of not having these mica plates, we shall have to 
 use a luminous source of heat and cut off the luminous 
 portion by means of a solution of iodine in bisulphate of 
 carbon ; that will cut off entirely the luminous rays and 
 leave only the dark ones, because this solution is absolutely 
 opaque and does not allow any luminous rays to pass 
 through it, but it permits the ultra red rays to pass through 
 it with great facility. Previously to placing that in the 
 beam I will place this thermopile as nearly as possible in 
 the focus of the rays. You see a deflection produced in the 
 galvanometer by the luminous rays falling upon it. I will 
 now cut them off, leaving only the dark field. The solution 
 
RADIATION. 189 
 
 is now interposed, and the only radiant rays which can pass 
 through are the non-luminous ones. I bring the needle of 
 the galvanometer to rest, and now we have a dark field, 
 allowing quite complete polarisation of the light ; but 
 when I turn the Nicol prism round until I see the 
 luminous image upon the near side of the iodine, which 
 will be absorbed, and not allowed to appear on the thermo- 
 pile, you will see that the needle is gradually deflected. 
 The field, which is a bright field, is also a hot field, and 
 when we turn the crystal round so that we have a, dark 
 field, we have also a cold field at the same time. Now we 
 have an intense amount of heat getting through ; by turn- 
 ing round the dark field you will see that the needle has 
 returned, and we shall have the cold field back again. 
 When it has returned, I interpose between the two prisms 
 the piece of mica which you saw produce so remarkable an 
 effect in the case of light. This experiment is a little more 
 delicate, but still I hope we shall be enabled to get a 
 sensible amount of heat allowed to pass through just as 
 we saw a very sensible amount of light pass through 
 before. You see the needle has moved through a consider- 
 able angle since the mica was placed there. Then we will 
 remove the mica, and you will see it return to zero. These 
 experiments were followed by a number of others ; and by 
 varying these experiments with the mica polarisers and 
 rock-salt apparatus Principal Forbes proved nearly all the 
 elementary facts of the undulatory theory of light to be 
 true with respect to heat. Perhaps the most striking was 
 the circular polarisation of light. M. Fresnel, by a most 
 wonderful application of mathematical analysis to physics, 
 viz., by an interpretation which he gave to the remarkable 
 mathematical symbol \/ - l, which had been found in some 
 researches of Sir Wm. Hamilton's and others to have a 
 peculiar significance, and by interpreting the significance of 
 the symbol in a peculiar way, was led by the equation 
 he had deduced to a most remarkable discovery, the circular 
 polarisation of light by what is called Fresnel's rhomb, 
 which consists of a piece of glass of a peculiar shape. It 
 is found that when light, polarised in a certain plane, falls 
 on one of those surfaces, and, after two reflections, is 
 allowed to pass out again, the light is in the peculiar con- 
 dition which is called circular polarisation, and the physical 
 
190 LECTURES TO SCIENCE TEACHERS. 
 
 meaning of that is this that supposing a ray of light is 
 passing from me, the whole of the ether which is set in 
 vibration is not vibrating backwards and forwards in any 
 particular direction, as is usually the case, but it is set 
 in a circular mode of vibration. As I have already said, 
 Principal Forbes found that by employing a Fresnel's 
 rhomb, made of rock-salt, at the proper angles, he was 
 able circularly to polarise heat also. This and several 
 other experiments I should have liked to show you, but 
 unfortunately I have none of my apparatus here. For this 
 one I am indebted to the kindness of Professor Guthrie, 
 but the result remains the same, that the application of 
 the undulatory theory to heat has been absolutely proved. 
 
 Perhaps the only connecting link that was wanted has 
 been supplied in late years to show that wherever there is 
 light there is also heat. The stars radiate to us a large 
 quantity of light, and if we could show that the stars 
 radiate heat to us, it was said by some that this would be 
 a final and conclusive argument of the identity of heat 
 and light. This was taken up by several astronomers, 
 Mr. Stone at the Royal Observatory at Greenwich, Mr. 
 Huggins, and also Lord Rosse. They all found effects of 
 this kind ; and Lord Rosse, by means of the apparatus 
 which you see before you, and by two thermopiles used in 
 a differential manner, was able to compare the heat 
 radiated by the moon and the space round the moon, and 
 at the different phases of the moon, by which he was led 
 to very remarkable results ; so that the fact has been 
 abundantly proved that both the stars and the moon 
 radiate a large amount of heat to us. 
 
 I wish to point out to you with regard to this diagram 
 that the form of the curves is entirely hypothetical. You 
 must not think that they are at all likely to be exact, 
 because we really know very little about the amount of 
 radiation from different parts of the spectrum ; it depends 
 very much on the absorption of the substance we employ. 
 Very likely the maximum ought to be moved closer to the 
 red end, but that is a question which we cannot settle 
 definitely. 
 
 I have attempted in a very feeble way to pass through 
 the various researches which led chiefly during the first 
 half of the present century to the final identity of light 
 
EADIATION. 191 
 
 and heat, and we now know with certainty that the radia- 
 tion of heat and the radiation of light are absolutely 
 identical, and that the only difference between them is in 
 the absorptive power of the substances composing our eyes 
 that they only allow a certain amount of rays to pass ; 
 there are rays not only beyond the red end but also beyond 
 the violet end of the spectrum. Those were first discovered 
 by Bitter, and this was investigated extensively by Professor 
 Stokes in a manner which I hope you will have the pleasure 
 of hearing from his own mouth in one of his lectures in 
 his celebrated experiments on fluorescence. 
 
 Of late years no experiments of great importance have 
 been made adding in any manner to this theory, but the 
 science of radiation has been advancing nevertheless. I 
 will simply allude to two pieces of apparatus here. This 
 is one used by Professor Tyndall in his researches on the 
 radiation of gases, or rather on the absorption of gases ; 
 but since radiation and absorption are equal, it follows 
 that where we have radiation we have an equal amount of 
 absorption. He employed a non-luminous source of heat. 
 He closed his tube with pieces of rock-salt, heat was 
 radiated through the tube, and struck on this thermopile. 
 The thermopile was in connection with a galvanometer, by 
 means of which the effect was measured. In the most 
 delicate experiments it is necessary to heat also the other 
 side of the thermopile, so that the mode of experimenting 
 is this : The tube is exhausted of air entirely, and the 
 current is reduced to zero by placing a vessel of hot 
 water at the other side, so that this side of the 
 thermopile is as much warmed as the other side is by the 
 radiation from the tube. A cube filled with boiling 
 water is placed at the other end of the tube : it absorbs a 
 certain amount of radiant heat, and this difference is 
 shown by a diminution in the temperature of this side of 
 the thermopile, which is indicated by the galvanometer. 
 The most interesting result of this was that apparently 
 gases which are of the most complex molecular structure 
 had the greatest effect in absorbing the red end of the 
 spectrum. Unfortunately in employing aqueous vapour 
 there is a tendency to deposit it on the sides, and you see 
 that not only is the heat radiated directly through this 
 tube, but also to the sides, and receives a number of 
 
192 LECTURES TO SCIENCE TEACHERS. 
 
 reflections from the sides ; consequently when aqueous 
 vapour was employed it was deposited on the sides of the 
 tube and diminished the radiation. It seems that this is the 
 explanation of the very large diminution in the amount of 
 heat transmitted when aqueous vapour was employed in 
 this tube. The celebrated Dr. Magnus employed a dif- 
 ferent form of apparatus, and also repeated this experiment, 
 and he came to the conclusion that aqueous vapour is not so 
 enormously absorptive of the ultra red rays as Professor 
 Tyndall found ; but in other cases Dr. Magnus's results 
 were in complete accordance with Professor Tyndall's. 
 
 In conclusion I would draw attention to the apparatus 
 of Professor Balfour Stewart for determining the radiation 
 of the sun at any one observatory day after day. It consists 
 simply of an iron case which has a motion in azimuth and 
 altitude, into which a thermometer can be screwed. This 
 pointed directly towards the sun, and the sun's heat is 
 concentrated, by means of a lens, upon the bulb in the 
 interior of the iron case. I have not time to explain the 
 principle of it, but in all probability this method will give 
 the most constant and most clear method ever devised for 
 measuring the amount of radiant heat from the sun at 
 different times. This is a problem which has always 
 offered the greatest difficulty, and a great deal of credit is 
 due to Professor Balfour Stewart for this invention, which 
 will probably lead to more accurate measurements than 
 have yet been obtained. The subject of radiant heat is so 
 extensive that it would have been interesting to have 
 devoted several lectures to the subject, but I have at- 
 tempted to-day to confine myself as far as possible to that 
 part which connects the theory of heat with the theory of 
 light. 
 
MICROSCOPES. 
 
 BY H. C. SORBY, F.R.S. 
 
 IN- tlie course of one hour it is quite impossible to do more 
 than give a very brief outline of the subject of the micro- 
 scope, for there are a great many departments connected with 
 it, each of which would well deserve to have a special lecture. 
 I might treat this subject on the present occasion in a variety 
 of ways. I feel very much tempted to go into it historically 
 on account of the very magnificent collection of microscopes, 
 of almost all periods, from the very earliest made down to 
 the most recent ; but I think it would be more useful to draw 
 your attention to those points which are the most important in 
 practically working with modern instruments. 
 
 "Without further introductory remarks, I will first call 
 attention to the reason of the magnifying power of a single 
 lens. If we have a plano-convex or double convex lens, 
 the rays proceeding from the focus are made parallel by 
 passing through the lens, and as our eyes are only constituted 
 to see distinctly when almost parallel rays enter them, the 
 result is that if an object be placed in the focus, and the eye 
 be placed on the other side of the lens, an object at that 
 short distance can be seen distinctly. If the lens were not 
 there the object could not be seen distinctly, unless placed at 
 a certain distance, dependent on the length of sight of 
 different individuals. Supposing that the distance at which 
 an individual can see distinctly is ten inches, and that the 
 lens has a focal length of one inch, an object clearly seen 
 in the focus will appear as large as one ten times the diameter 
 at the distance of ordinary vision, because it will subtend as 
 
 o 
 
194 LECTURES TO SCIENCE TEACHERS. 
 
 large an angle. The magnifying power of a single lens thus 
 depends on its "bending and making more or less parallel rays 
 that would otherwise enter the eye at a very divergent angle. 
 We may indeed obtain a similar apparent enlargement by 
 looking through a minute hole in a blackened card, which 
 cuts off the divergent rays, but in this case so much light is 
 lost, and the diffraction is so great, that the definition is bad. 
 We thus see that the so-called magnifying power of simple 
 lenses is due merely to the fact of their enabling us to see 
 objects distinctly when they are very near to the eye. 
 
 Single lenses are useful for examining an object with com- 
 paratively low powers, but they do not give perfect definition 
 unless they are only very small segments of spheres and none 
 of the rays are very much bent. They do not bring all the 
 rays to an exact focus, but have what is called spherical 
 aberration. The rays of light passing near the centre are 
 brought to one focus, those passing further from the centre 
 are brought to a focus nearer to the lens, and those at the 
 extreme edges still nearer. The result is that in a single lens 
 made with spherical curves you never can have an object 
 exactly in focus, since you have different focal lengths for 
 different parts of the lens. That could be overcome by 
 lenses with elliptical or hyperbolic curves, which would bring 
 all the rays to one focus, but the difficulties of making them 
 would be so great that, practically speaking, they cannot be 
 made in a satisfactory manner. I draw your attention to 
 the fact because the so-called spherical aberration with which 
 we have to contend in making microscopes is a matter of 
 very great importance, and it is very desirable that the true 
 source of the difficulty should be known. 
 
 Various combinations of lenses have been made in order to 
 avoid as much as possible this spherical aberration. It would 
 occupy too much time to enter into details, but by combining 
 different lenses the spherical aberrations may be so greatly 
 reduced that very valuable doublet and triplet lenses may be 
 constructed, which are exceedingly useful for so-called simple 
 microscopes that is to say, microscopes having only a single 
 system of lenses. The chief advantage of simple microscopes is, 
 that the object is seen in its proper position it is not inverted. 
 In dissecting and in preparing objects they are very useful, 
 because all the motions of the hand are seen in the natural 
 direction. We are, however, limited very much in their use 
 
MICROSCOPES. 195 
 
 since, in order to get very high magnifying powers, the focal 
 length must be very short, and then the objects are so very 
 close to the lens that it is almost impossible to perform any 
 necessary operations under them. 
 
 Independently of the so-called spherical aberration, we 
 have to contend with what is called chromatic aberration. 
 As I have said, the spherical aberration might be overcome if 
 it were possible to grind elliptical or hyperbolic lenses ; but 
 even then we should still have to contend with what is called 
 chromatic dispersion. When light is passed through a prism 
 with inclined faces, it is more or less bent out of its original 
 course ; but the different rays are thus refracted to a different 
 extent, the blue rays being much more bent than the red rays 
 and the green to an intermediate extent. The result is that 
 there is a different focal point for the different rays of the 
 spectrum. When white light passes through any ordinary 
 simple magnifying lens the focus for the blue is thus shorter 
 than for the red. Hence, on looking at an object, you see it 
 fringed with colours, and get an indistinct image, and you 
 see colours that ought not to exist. At one time it was 
 thought that the evil could not possibly be remedied ; and 
 even so great an authority as Sir Isaac Newton concluded that 
 it could not be overcome. Fortunately, however, we have 
 now the power of overcoming it to a very great extent. The 
 principle on which this depends is, that the extent to which 
 the different rays are separated by different kinds of glass 
 their so-called dispersive power does not vary directly as 
 their refractive power, and it is thus possible to combine 
 together two different kinds of glass, so that the dispersive 
 power of one may counteract the dispersive power of the 
 other, and yet only partially counteract its refractive power. 
 This is best illustrated by what is seen in compound prisms 
 made of crown and flint glass, the relative dispersive power 
 of which latter is much greater than that of the former. 
 According to the angles of the two kinds of glass, prisms may 
 be so combined that we may obtain dispersion without refrac- 
 tion, or refraction without dispersion, the former being a 
 so-called direct-vision prism, through which the light passes 
 in a direct line, but much dispersed, and the latter an 
 achromatic prism, bending the light but not dispersing it, and 
 thus showing objects out of the direct line, but free from 
 coloured fringes. This is what we want in constructing 
 
 o 2 
 
196 LECTURES TO SCIENCE TEACHERS. 
 
 microscopes. Success depends almost entirely on the property 
 I have described, and on compounding together flint and 
 crown glass, in such a manner that you can take advantage 
 of a considerable amount of refractive power, but almost 
 entirely obliterate the dispersive power. This may be done 
 by combining a double convex lens of crown glass with a 
 plano-concave lens of flint glass, and when the surfaces have 
 the proper curves, and the lenses are of the proper thickness, 
 the dispersive power of the crown glass may be counteracted, 
 whilst its refractive power is only partially reduced. We 
 can thus construct lenses of moderately short focus length, 
 in which the images of different rays are brought to the same 
 focus, and thus can obtain by this kind of combination 
 simple achromatic lenses. 
 
 I will now proceed to the consideration of compound 
 microscopes, because these are in the present state of science 
 far the most important. The characteristic principle of these 
 consists in their having an object-glass placed at the lower 
 end of the body of the instrument, which forms at the upper 
 end an image of any object on the stage. This image is itself 
 considerably larger than the object, the diameter being as 
 many times greater as the length of the tube from the object- 
 glass to the image is greater than the distance from the object- 
 glass to the object on the stage. The image thus formed is 
 inverted both vertically and laterally, and Avould appear per- 
 fectly distinct if it were thrown upon a white surface and 
 examined in front, or on ground glass and seen from behind. 
 If no such screen be placed to receive the image, it is still, as 
 it were, formed in free space and is capable of being again mag- 
 nified by another lens, or system of lenses. The principle of 
 compound microscopes consists in their forming an enlarged 
 inverted image by means of an object-glass, and in making it 
 visible when near to the eye, or, so to speak, magnifying it 
 with an eye-piece. 
 
 In the very earliest compound microscope this eye-piece 
 consisted of a plano-convex or double-convex lens, but it was 
 soon found that a combination of two plano-convex lenses gave 
 a far better result. Modern eye-pieces are of two kinds. The 
 Huyghean eye-piece consists of two plano-convex lenses the 
 tipper one called the field-lens and the lower the eye-lens, 
 with a diaphragm between them at the focal point of the eye- 
 lens. The eye-lens is principally concerned in magnifying 
 
 
MICROSCOPES. 197 
 
 the image, and the field-lens in increasing the field. The 
 other eye-piece is called the negative, or Kelner's eye-piece, 
 and has no diaphragm, and the focus of the upper lens is on the 
 field- lens. The advantage of this is that we get a much larger 
 field, but the disadvantage is that the definition is not so good. 
 
 Since we thus examine with the eye-piece an inverted image 
 of the object, it is in all cases seen as it were turned upside 
 down. This general principle of magnifying an inverted 
 image was brought into use at as early a period as 1590, by 
 Janssen, whose most interesting instrument is in the Exhi- 
 bition, though, in comparison with our modern microscopes, it 
 might be looked upon as little better than a toy. 
 
 The magnifying power of compound microscopes depends 
 on both the object-glass and the eye-piece. The shorter the 
 focal length of the object-glass the larger is the image formed 
 in the upper part of the instrument. The final result also 
 depends on the magnifying power of the eye-piece itself. We 
 can thus increase the magnifying power of the instrument 
 by having the object-glass of shorter focal length ; and can 
 likewise increase the power by using eye-pieces of shorter 
 focal length. On the whole it is, however, better to use 
 an object-glass of short focal length, and an eye-piece of 
 moderate power, than to use an object-glass of long focal 
 length and enlarging the image with an eye-piece of too 
 great power, since we can thus utilize a greater beam of 
 light, and obtain better definition. 
 
 The simplest form of achromatic object-glass that could be 
 used would be a combination of a double convex lens of crown 
 glass, with a plano-concave of flint ; but though one of this 
 kind may give a good result when the focal length is con- 
 siderable, it would be impossible to obtain good definition 
 when the focal length is small, and the aperture great. By 
 combining three such compound lenses of different sizes 
 somewhat differently corrected, the spherical aberration may 
 be very greatly reduced, and the general effort much improved, 
 since a short focus can be obtained with lenses of less curva- 
 ture than when only one compound lens is used. Until the 
 last few years most of the best object-glasses were constructed 
 on this principle. Some modern high- power object-glasses 
 are, however, much more complex, and consist of a combi- 
 nation of as many as eight lenses, nearly all of different 
 curvature or size. The magnifying power is in many 
 
198 LECTURES TO SCIENCE TEACHERS. 
 
 combinations chiefly due to the lenses placed at the "bottom 
 of the object-glass, and the spherical and chromatic aberrations 
 are corrected as far as possible by the lenses placed above 
 them, but as I do not suppose the ladies and gentlemen here 
 present are at all likely to make microscopes for themselves, 
 I will not occupy time by going into all the details of con- 
 struction of object-glasses. It is a most difficult subject, both 
 in theory and practice, and the more one knows of the diffi- 
 culties to be contended with in making lenses of high power 
 the more one feels astonished at the excellence of some that 
 are now made. Some of the high-power object-glasses, with 
 lenses as small as a pin's head, are all so accurately finished, 
 fitted, and adjusted, that in my opinion they are triumphs of 
 science and skill. 
 
 In looking at an object with a high-power object-glass 
 made perfectly correct, both for chromatic and spherical 
 aberration, if you place over the object a piece of thin glass 
 with water or Canada balsam underneath, the corrections are 
 no longer strictly accurate, and it is necessary to make 
 different corrections in the object-glass. This is accomplished 
 by means of a screw in the collar which is made to turn round, 
 and the lenses can be made to approach to or recede one from 
 another. In this manner corrections can be made for the 
 difference in the conditions due to the glass cover. 
 
 Many object-glasses are what are called dry lenses, that is 
 to say, there is air between the object-glass and the object. 
 In low powers this is almost invariably the case, but in some 
 of the high powers now made, advantage is taken of what is 
 called the immersion principle. In using such object-glasses a 
 small quantity of water is placed between the front lens and 
 the glass cover. One great advantage of this is that we get 
 a much greater amount of light. Another advantage is that 
 you can correct the lens more easily, and perhaps utilize a 
 wider beam of light. It would be tedious to enter into the 
 discussion that has taken place with reference to the difference 
 in the size and aperture in dry and immersion lenses. Many 
 most eminent authorities have differed exceedingly about this, 
 and a most angry discussion has taken place on the subject. 
 In any case, it appears that with immersion lenses we obtain a 
 greater amount of light, especially that of wide deviation from 
 the direct line of vision, which is of the greatest value in 
 defining minute structure. 
 
MICROSCOPES. 199 
 
 I will now proceed to say a word or two with regard to the 
 construction of the instrument itself, and the general mechanical 
 arrangements requisite to hold the optical part and the object 
 under examination, and for illuminating it properly. The 
 Exhibition contains as simple a compound microscope as is 
 possible, made by Janssen, the first ever constructed. There is 
 no stage and no mirror to reflect the light, merely a simple 
 lens for 'object-glass and a single lens for eye-piece, held 
 in tin tubes. There are also many kinds of most inter- 
 esting microscopes, of nearly all periods down to the most 
 modern, but I will allude to only a few to illustrate some 
 particular points. Passing down from the one just named 
 by Janssen, we come to those forms of instrument which are 
 especially characterized by very large eye-pieces of two lenses 
 on Huyghens's principle. As an example, I may refer to the 
 two microscopes made by Marshall in the commencement of the 
 eighteenth century, the eye-pieces having field-lenses of two 
 inches in diameter, which makes the instrument look most 
 remarkably stout. Few microscopes have been made larger 
 or more complex than those by Martin in the middle of the 
 eighteenth century. One of these in the Exhibition belongs 
 to the Royal Microscopical Society, and it was thought to be 
 the only one in existence, but another has turned up which 
 I had the opportunity of putting together ; I must confess 
 I had very great difficulty in finding out where the different 
 things fitted. There is almost every movement that any one 
 could possibly devise, and a great many of them for the 
 practical work of the present day are not only useless but far 
 worse than useless. Then again the construction is exceed- 
 ingly curious. One tube goes into another for only about one- 
 eighth of an inch ; and there is scarcely a firm joint in the 
 whole apparatus. There are movements of nearly every 
 decription ; you can move the body of the instrument up 
 and down, from side to side, and backwards and forwards, 
 but there is scarcely a single joint which one could call 
 good. These instruments may be looked upon as the 
 opposite extreme of that by Janssen. 
 
 Leaving the microscopes of mere historical interest and 
 coming down to modern times we have examples of the 
 most simple and complex arrangements. Some are of the 
 very simplest form, more in vogue on the Continent than in 
 this country ; so constructed that you have not the power of 
 
200 LECTURES TO SCIENCE TEACHERS. 
 
 inclining the body, but have to look straight down. There 
 is a disadvantage in that because the tear of the eye runs 
 over the pupil, and you do not get such good definition as 
 when you can incline the microscope to any convenient angle, 
 so that the tears may not remain in front of the pupil. 
 Some have the stage as simple as it is possible to be, the 
 body is a simple tube, and you get a coarse adjustment by 
 pulling it up and down. There is a reflector below the 
 stage, but scarcely anything which is not absolutely requisite. 
 
 Many of the microscopes in the Exhibition have various 
 excellent and valuable movements. It is very important to 
 have no vibrating motion in the tube, because if you are in a 
 place where there is the least tremor, and the tube is able to 
 move in the least degree from side to side, the object is seen 
 to vibrate in the most inconvenient manner. Some of the 
 early microscopes are for this reason totally unfit for difficult 
 observations. I think it a disadvantage to have the body 
 move in any direction but up and down, unless the instrument 
 is very well turned out of hand. If it is made to perfection the 
 power of turning on one side may be very good for particular 
 purposes. An excellent stand for holding the tube is the 
 so-called Jackson model. The body, the stage and the sub- 
 stage, are all firmly held by one solid piece of metal, and must 
 therefore vibrate together in such a manner as to be of little im- 
 portance. The body slides up and down on a long groove with 
 rackwork, and there can be no more secure means of prevent- 
 ing lateral movement, because it is held so firmly for such a 
 distance. 
 
 Proceeding to the stage, we have some of the most simple 
 form, and others with very complex movements, so that you 
 can move the objects up and down or from side to side by the 
 milled heads, or can rotate the whole round quickly or slowly 
 with the screw, the object remaining all the while in the 
 centre of the field. I might say much more on these various 
 arrangements if time would permit, but cannot, I think, do 
 better than express my general conclusions in a simple form. 
 I would strongly insist on the desirability of having every 
 accessory movement well made or not made at all. Whatever 
 movements there are should do only just what is required, 
 and should admit of no other motion, and this can only be 
 accomplished by good and careful workmanship. If you do 
 not wish to be at the expense of good workmanship you 
 
 
MICROSCOPES. 201 
 
 should have a microscope of simple construction. It is 
 far better to have one of the simplest form with very few 
 movable parts, than to have a microscope with all the 
 movements 1 have described badly executed. If you want a 
 cheap microscope, have a simple one, but do not get a cheap 
 microscope which has a great many movements which work 
 irregularly, give tremor, or throw the object out of focus. 
 
 I now come to the practical use of the instrument. Assuming 
 that you have a microscope of a satisfactory kind, when you 
 come to use it success depends as much on properly under- 
 standing how to illuminate the object as on the quality of the 
 instrument itself. You may have a splendid microscope, but 
 unless you illuminate the object properly, you will see nothing, 
 for the simple reason that there is nothing to see. I think I 
 could not do better on the present occasion than deal some- 
 what at length with this part of my subject, since it is of 
 such great importance in the practical use of the microscope. 
 First of all then, I will draw your attention to the surface 
 illumination to which we must have recourse in examining 
 opaque objects, and which may sometimes be used with ad- 
 vantage in the case of objects which are not opaque. With 
 this kind of illumination very much depends on light and 
 shade, as in the cases of objects which we see with the naked 
 eye. We all very well know that if you have light thrown 
 directly on a wall it may appear to have a perfectly uniform 
 surface ; but if the light be thrown very obliquely all the 
 irregularities of the plaster, blisters and such things, are made 
 very conspicuous. This illustration serves to show that if 
 you have the light thrown at a certain angle you can see no 
 character whatever ; but if thrown at another angle the surface 
 structure may be seen as well as could be desired. In some 
 cases if you were to throw the light at too oblique an angle 
 certain structures might be hid by the shadows of other 
 greater irregularities, so that for each particular object it is 
 very important that you should throw the light at particular 
 angles, and in examining an unknown object it is desirable to 
 try the effect of light of various angles of obliquity. There 
 are various ways of obtaining this surface illumination. We 
 can have a bull's-eye condenser throwing the light at various 
 angles from the side, or we can use a silvered parabolic reflector, 
 the first of which was made for my own use in examining iron 
 and steel. One advantage of these parabolic reflectors is that 
 
202 LECTURES TO SCIENCE TEACHERS. 
 
 the shadows appear true. When you examine an object with 
 a bull's-eye condenser throwing light from one side to the 
 other, you instinctively know which way the light is coming ; 
 and when you look at an object, the object being inverted, 
 you see the lights and shadows all the wrong way. On the 
 contrary, in using these parabolic reflectors the light is thrown 
 from the opposite side, and the image being inverted by the 
 microscope, when you look at the object the lights and 
 shadows appear natural, as they would do if you looked at the 
 object without a microscope. That is certainly a great advan- 
 tage ; but there is this disadvantage in a fixed reflector, that 
 you cannot alter the angle at which the light is thrown. 
 There is, however, no reason why we should not so modify 
 these reflectors, and so arrange them with reference to the source 
 of light, as to partially overcome this objection. We have 
 also what are called Lieberkuhns, which are sections of a 
 concave silvered mirror. The light is sent up by the mirror 
 at the bottom of the microscope, and reflected back on all sides 
 of the object by this concave mirror to the object, which is thus 
 equally illuminated in all directions. There is a hole in the 
 centre of the mirror to admit the light to the object-glass, so 
 that there is no absolutely vertical light. In some cases it is 
 an advantage thus to have the light equally in all directions, 
 since it enables you to see certain structures well This is, 
 however, sometimes a disadvantage, since it gives no shadows, 
 but this could be overcome by stopping out portions of light 
 from one side or the other. With a Lieberkuhn you usually 
 do not see structures revealed by light and shade, but see 
 those due to such difference of texture as reflect light in a 
 varying manner. If the object is very small or transparent 
 you must use what is called a dark well put immediately 
 under it, so as to prevent the light passing at the side or 
 through it. With this illumination perfectly transparent 
 substances appear black, but if any part of the object be full 
 of little colourless granules they look white, and if coloured 
 the colours are well seen. These kinds of surface illumination 
 cannot be used with very great advantage, except with mode- 
 rately low powers, but for them they are often everything that 
 could be desired. 
 
 The bull's-eye, parabolic mirror, and Lieberkuhn show objects 
 illuminated on the upper side by more or less direct reflec- 
 tion ; but another class of illuminators which like them give a 
 
MICROSCOPES. 203 
 
 naturally dark field, show the structure of the object "by light 
 thrown on them from below, bent out of its normal course by 
 more or less oblique reflection and refraction. This may be 
 accomplished in various ways. The light may be thrown on 
 the object by the mirror from one side in such a manner that 
 if no object were on the stage it would not pass at all up the 
 body of the microscope. But if the object placed on the 
 stage is granular, or has little facets and other irregularities 
 of structure, the light would be so bent as to pass up to the 
 eye. This may be accomplished also by means of two or 
 three different kinds of illuminators, such as Amici's prism, 
 Mr. Wenham's reflex illuminator, and his parabolic reflector, 
 or by a large lens of short focal length, with a central black 
 spot to stop out direct light. All these different kinds of 
 illumination depend upon the light being thrown so obliquely 
 from the under side that it will not enter the object-glass 
 unless it be turned out of its course by the reflection or the 
 refraction of the various parts of the object. One great 
 advantage is that the field of the microscope is dark, and the 
 eye is not in any way distressed by the light which comes in 
 at the sides. Certain kinds of structure are also seen to very 
 great advantage, especially with binocular instruments. I am 
 inclined to believe that the further development of this kind 
 of illumination would yield better results than can be obtained 
 by the ordinary method of illumination by transmitted 
 light. 
 
 Another kind of illumination which is on the whole more 
 common, is whers the light is reflected directly up the tube 
 of the microscope, so that if no object be placed on the 
 instrument the field of the microscope is filled with light, 
 and if any object be there it is seen by variation in the 
 intensity or colour of the light. I wish to draw your 
 attention to one or two points connected with this kind of 
 illumination, because, in my opinion, its further development 
 with certain modifications which I believe have not been 
 carried out, will perhaps enable us to overcome certain 
 difficulties which at present stand in our way. Much may 
 be learned by the study of mineral structures, since in the 
 case of crystals and of solid portions of glass and other 
 analogous objects, we know what their character is, whereas 
 in the case of minute organic structures we have rather to 
 infer what is their structure from what we see. Therefore in 
 
204 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 forming some general idea of illumination, I think we may 
 learn a great deal by studying what we see in small crystals, 
 and in inorganic bodies of pretty well known form. If light 
 be sent straight up through a small crystal, having parallel 
 faces terminated by an oblique plane, it will pass up directly 
 in one part and be bent .and thrown aside altogether in the 
 other, and in looking at such a crystal with the microscope, 
 you might see a black edge with perfect definition, but if you 
 were to throw the light obliquely with a condensing lens of 
 considerable aperture, the light might enter at such an angle 
 that it would be bent and pass straight out, and if the edge 
 of the crystal were absolutely perfect you would see no dark 
 band at all. When thus illuminated, the crystal might be 
 quite invisible. 
 
 Ft G./. 
 
 FIG. 2. 
 
 Ml 
 
 My meaning will be better understood by means of the 
 following rough illustrations. 
 
 In Fig. 1 the light is supposed to be all parallel, and is 
 bent quite out of the line of vision by the oblique end of the 
 crystal, whereas in Fig. 2 the light is supposed to be so very 
 variably divergent, that some of it can enter the object-glass 
 after passing through the crystal, both where the sides are 
 parallel and where they are inclined. Such a case clearly 
 
MICROSCOPES. 205 
 
 shows the importance of regulating the aperture of the con- 
 densing lens used for illumination, since with one aperture 
 the edge of the crystal would be shown by a dark band, and 
 in the other might be invisible. When low powers are used 
 the most convenient condenser is a moderately large plano- 
 convex lens of short focal length, or two so combined as to 
 give less aberration, but in using high powers an achromatic 
 condenser of more complex structure, short focus and 
 large aperture, is very desirable. Even with rough and im- 
 perfect mechanical arrangements I have been able to see 
 sufficient to convince me that even with achromatic con- 
 densers it is a very great advantage to have two different 
 diaphragms, one to modify the divergence of the light, and 
 the other the size of the beam passing up through the sub- 
 stage. By using only light of high angle of divergence, and 
 a small opening in the sub-stage diaphragm, some objects are 
 seen to very great advantage. Such an illumination is im- 
 possible with condensers constructed in the ordinary manner, 
 with only one diaphragm placed in such a position that it 
 gives imperfectly both these effects combined. 
 
 You must remember that the light passing near the centre 
 of a condenser is inclined at a small angle to the line of 
 vision, whilst that which passes through the exterior zone 
 is much more divergent. Hence the light from the outside 
 part of the condenser may pass through certain parts of the 
 object, but if you stop off the outer zone of the condenser 
 no light may pass through those parts and you may see 
 a well defined dark edge in such a crystal as I have alluded 
 to. By making the opening wider and wider, and thus 
 allowing more and more divergent light to pass to the 
 object, you may as it were obliterate the crystals on the 
 stage. It is sometimes important thus to be able to make 
 the light pass through an inclined edge. For instance, I 
 have a beautiful crystal of sapphire with a fluid cavity con- 
 taining liquid carbonic acid. It happens to lie in such a 
 position in the crystal that if I use light of a moderate angle 
 of divergence, the fluid cavity is completely hidden, because 
 no light passes through where it lies, but on increasing the 
 aperture and the obliquity of the light the dark shadow dis- 
 appears, and the fluid cavity is perfectly well seen by trans- 
 mitted light. The possibility of either seeing the object or 
 not seeing it at all, thus entirely depends on knowing 
 
206 LECTURES TO SCIENCE TEACHERS. 
 
 how to modify the angle of divergence of the light used for 
 illumination. If you had light passing through parallel, like 
 that reflected from an ordinary direct mirror, you would see a 
 broad dark band, and would not have the remotest idea that 
 such an interesting cavity existed ; but if you used as a con- 
 denser, an almost hemispherical lens, you would then see the 
 fluid cavity to perfection. The use of such simple single or 
 compound condensing lenses of large aperture is not common, 
 but they enable us to study certain objects in a very satis- 
 factory manner, since with appropriate stops the angle of 
 divergence of the light is so very much under control. This 
 principle may also be brought to bear in another class of 
 objects. There may co-exist two perfectly independent 
 kinds of structure, which with ordinary illumination may so 
 far unite together as to produce a general appearance of a 
 very misleading character ; but by varying the divergence of 
 the light or the size of the sub-stage opening, first one and 
 then the other structure may be made invisible, whilst the 
 other is seen to great advantage by itself. 
 
 Another interesting illustration of the importance of the 
 angle of divergence of the light is furnished by the little 
 spherical cavities met with in amber, some filled with gas 
 and some filled with water. By proper illumination you can 
 see these very well, but I find that the character of the object 
 and the illumination depends on a set of conditions which 
 I believe have not attracted any attention. In looking at 
 such spherical cavities in minerals, which I choose as a type 
 because they are so simple, and we can understand them 
 perfectly well, or in looking at, a spherical cavity or a bubble 
 in glass, you see first a black outline and then a white bright 
 centre. This bright centre is nothing more than the image 
 of the opening at the bottom of the condenser seen out of 
 focus, and by a little modification in the focus you can see 
 very distinctly anything special in its character. My 
 meaning will be better understood by giving a few illustrations 
 of what may be seen in examining the gas cavities, which are 
 so extremely numerous in some specimens of amber. When 
 the opening in the sub-stage diaphragm is large, you see a 
 broad clear space and a relatively narrow dark band, extending 
 inwards from the external outline of the cavity, as shown by 
 Fig. 3 ; but on making the opening in the diaphragm small, 
 this black zone closes in, and there is only a small bright 
 
 
MICROSCOPES. 207 
 
 centre, as in Fig. 4, which you can see distinctly is nothing 
 but the image of the hole in the diaphragm, seen more or less 
 in focus according to the adjustment. If you take a diaphragm 
 with a central stop, you see what looks very strange, viz., a 
 bubble or spherical cavity with a central black spot, from 
 which proceed the two small arms, as shown by Fig. 5. This 
 appearance is thus manifestly due to nothing but the image 
 
 O 
 
 FIG. 3. FIG. 4. FIG. 5. 
 
 of the opening in the diaphragm, the central stop, and the 
 two arms which support it. These very simple facts thus 
 clearly prove that what might easily be mistaken for structure 
 may be merely the light and shade depending on the kind 
 of illumination made use of. This way of viewing the sub- 
 ject has occurred to me only quite recently, and I feel 
 persuaded that an arrangement which gives us the means of 
 limiting the obliquity of the light and also the size of a 
 more distant opening is very useful, and would be equally 
 applicable in the case of rods, and minute fibres, and such 
 kinds of structures as are commonly met with in organic 
 bodies. In the case of the markings on some diatoms, we 
 do indeed see all the appearances that would be due to a 
 vast number of small lenses arranged side by side ; and, as 
 in the case of the above-described cavities, we may so modify 
 the illumination as to show the external outlines of these 
 bead-like lenses, or to see merely a central black spot in each 
 of them. As a general rule, however, the curvature of the 
 surfaces of organic bodies are seldom sufficiently regular to 
 give distinct images of the openings in the sub-stage 
 diaphragm, but yet there can be little doubt that the lights 
 and shades are to a very great extent due to a similar 
 cause, and that as far as this is concerned they might be 
 looked upon as very much distorted and irregular spherical 
 or cylindrical lenses. In any case, much more may be learned 
 of their true form by carefully observing and discussing the 
 changes that take place, on varying the divergence of the 
 
208 LECTURES TO SCIENCE TEACHERS. 
 
 light and the diameter of the sub-stage opening, than can be 
 actually seen with any one single illumination. What I am 
 anxious to do is by an inductive process to be able by further 
 examinations to form something like an intelligible ex- 
 planation of the cause of the lights and shadows that we see 
 in objects as examined by transmitted light, and in order to 
 do this it is most important to be able to vary the illumina- 
 tion very much, and at the same time to know exactly what 
 its true character is whether slightly or very divergent, or 
 proceeding from a small or wide sub-stage opening. 
 
 Independently of objects whose structure is shown by lights 
 and shadows in the way I have alluded to, we have many 
 which are seen by the difference in their colour, arid then it 
 is only a question of illuminating the surface, or transmitting 
 light through the object. In sections of rocky substances, 
 the different constituents, black, red, brown, or otherwise, are 
 seen at once by their colour, independently of any light or 
 shade. This method of distinguishing objects by difference 
 in colour is extensively employed in studying organic struc- 
 tures, by using a staining on which they act like mordanted 
 textile fabrics. Various staining materials will thus combine 
 with certain constituents and colour them, and have a very 
 slight effect on others. In that manner you recognise the 
 different constituents of what otherwise would appear like 
 almost homogeneous structure. 
 
 Another important question connected with the microscope 
 is illumination by means of polarised light. I must take it 
 for granted that you all know what polarised light is, since 
 time would not permit me to describe its characters. One 
 principal point is, that polarised light has different properties 
 in different directions. We can sometimes make use of it 
 very effectively independent of an analyser, by having a 
 polariser under the stage of the microscope and illuminating 
 the object with polarised instead of with ordinary light. 
 Polarised light, when the plane of polarisation is inclined in 
 certain directions to an edge of a crystal or other inclined 
 surface, may be to a great extent reflected and not transmitted 
 to the eye in cases where the amount reflected would be very 
 much smaller if you used ordinary light. In this manner 
 you may define crystals or other transparent objects mounted 
 in Canada balsam, which is of so nearly the same refractive 
 power that they can scarcely be seen with ordinary light. 
 
MICROSCOPES. 209 
 
 In such cases the use of polarised light without an analyser 
 may be very useful. You. use it simply as light, and rotate 
 the prism so as to get the plane of polarisation in different 
 directions. The more common plan is to use it with an 
 analyser placed generally over the eye-piece. By certain 
 adjustments, when no object is on the stage, the whole field 
 of the microscope is dark; but if you place under the 
 microscope objects which have more or less powerful double 
 refraction, they depolarise the light and appear either white 
 or coloured, dr invisible, according to the angle at which 
 they are placed in relation to the plane of polarisation. In 
 order to study these effects, either the object or the plane 
 of polarisation must be rotated. Time will not permit of 
 entering further into it, but I may just draw your attention 
 to one or two illustrations. Some objects may show little 
 or no structure when examined with ordinary light, but by 
 using polarised light various dark markings and colours may 
 be seen, which vary as you rotate the plane of polarisation. 
 By careful induction you may form a very accurate opinion 
 as to the kind of structure before you. Thus, if the object 
 give a well-defined black cross rotating with the plane of 
 polarisation, it indicates that small crystals radiate uniformly 
 from a single centre ; whereas, if there be an irregular varying 
 distribution of dark patches shading off into bright or coloured 
 portions, all changing gradually as the plane of polarisation is 
 rotated, there must be an irregular grouping of imperfectly 
 radiating crystals. On the other hand, if on rotating the 
 plane of polarisation the object becomes uniformly dark and 
 bright, you know it is a portion of one simple crystal. A 
 great many other most interesting facts may be made out in 
 that manner by using polarised light. Such general con- 
 clusions are more simple and obvious in the case of mineral 
 structures, but are by no means confined to them. Much 
 may thus be learned respecting the arrangement of the 
 mineral matter in the various kinds of calcareous shells and 
 other organisms, and I cannot but think that much remains 
 to be learned even in the case of more purely organic 
 structures. 
 
 When we use very high powers 1,000 linear, or upwards 
 and employ object-glass of very short focal length, we must 
 contend with another class of difficulties quite distinct from 
 any I have hitherto mentioned. In using very high powers 
 
 p 
 
210 LECTURES TO SCIENCE TEACHERS. 
 
 we are approaching, if we have not already passed, the 
 limit that is allowed to us by the physical characters of 
 light itself. When we have to deal with objects from 
 
 BO Q 00 
 
 the length of the waves of light, and must contend with 
 difficulties due to the fact that we are trying to examine objects 
 which are small in proportion to the waves of light. It is a 
 question, if it be not already certain, that even if the optical 
 contraction by transmitted light were in every respect perfect, 
 we could never distinctly see the true outline o"f objects less 
 than half a wave-length of light, though their mere presence 
 might be recognised. It may seem strange to say it, but the 
 fact appears to be that the physical constitution of light itself 
 is too coarse to enable us to see all that we could desire, and 
 could see if light were of a more refined character. A great 
 cause of this difficulty is what are called interference fringes. 
 If you have an object which ought to look like one dark 
 line, and examine it with a high power, you find dark and 
 coloured lines on each side, and in certain cases these are 
 almost as distinct as the object itself. There appears to be 
 little doubt that certain structures have been described by 
 microscopists which look as distinct as if they were real 
 markings, though due to nothing more than interference 
 fringes. Such an inaccurate interpretation is entirely due 
 to the fact that light, so to speak, breaks down when we 
 try to examine objects which are small as compared with 
 the waves of light. It would occupy a very long time 
 to enter into the full particulars, but I may say that one 
 of the principal means of overcoming this difficulty is by 
 increasing the aperture of the lenses. The width of these 
 interference fringes depends to a considerable extent on 
 the angle of divergence of the light used in illumination, 
 and passes into the object-glass, and by increasing the 
 aperture so as to bring into play more and more oblique 
 rays, the size of these fringes is lessened, and thus you 
 improve the performance of the instrument. By theory 
 the defining power ought to vary as the chord of the angle 
 of aperture, and microscopists working practically with- 
 out any regard to the undulatory theory of light had come 
 exactly to the same conclusion. With dry lenses you do not 
 get so good result as with the immersion lenses ; since with 
 the latter the interference fringe would be only three-fourths 
 
MICROSCOPES. 211 
 
 the width of those in the case of a dry lens of equal aperture, 
 and therefore, other things being equal, an immersion lens 
 may be said to have four-thirds the defining power of a dry 
 lens. In all these kinds of examinations, in order to fully 
 utilise the capabilities of the object-glass, the light passing 
 from the object ought to be of the same angle of divergence 
 as that which can enter the object-glass. With an object 
 not covered you might obtain greater divergence, if light 
 could be made to pass in at a greater angle. However, in 
 making object-glasses of very wide aperture we come across 
 great difficulties. By increasing the size of the aperture we 
 increase the difficulty in making them correct in other respects, 
 especially in correcting the spherical and chromatic aberrations. 
 Another practical difficulty is that the object-glasses come 
 down so close to the object, and it is impossible to see any- 
 thing through a thick cover, or when some distance within 
 a transparent portion of the object itself, as is so often the 
 case in studying the microscopical structure of minerals, or 
 the crystals enclosed in blowpipe beads. I very much fear 
 that it is only too true that if you improve object-glasses 
 in one respect, you make them worse in another. Those 
 qualities which are necessary for one purpose are unnecessary 
 for another ; and in my opinion we ought to have object- 
 glasses constructed for the particular kinds of work we wish 
 to use them for. If we wish to define very close markings 
 on thin flat objects, like diatomaceae or the minute striae 
 of muscular fibre, we must have a very wide aperture, even 
 if the object is then so very close to the object-glass that the 
 range of vision is too small when we come to study other 
 kinds of objects. For instance, in some cases, with the very 
 best object-glasses of a very high angle, which would give 
 the most splendid definition of test objects, you might not 
 be a"ble to see other objects you wanted to examine, because 
 you could not get at them. They would be beyond the focal 
 point when the object-glass touched the cover. In such a 
 case, with only moderately fine structures, an object-glass of 
 much less cost and smaller angle of aperture might enable 
 you to see all the necessary detail, and you could get at the 
 object, even when under one-tenth of an inch of Canada 
 balsam, or other transparent substance, where you could not 
 have seen it at all with a lens of high angle and of the same 
 magnifying- power. I am, therefore, inclined to believe that 
 
 p 2 
 
212 LECTURES TO SCIENCE TEACHERS. 
 
 the kind of object-glass ought to depend on the kind of work 
 we wish to do. If we want to examine the very minute 
 structure of thin flat objects we must use one kind, and if 
 we want to look at things through a considerable thickness 
 of water or glassy matter, and to study various irregular 
 objects met with in original research, we must have glasses 
 which do not approach too near the object, even if the angle 
 of aperture and the definition of very close markings are by 
 this means somewhat diminished. 
 
 In connection with the visibility of very minute objects 
 there is another interesting point. If in order to define objects 
 that are very close together you increase the angle of the 
 divergence of the light, you then bring into play another prin- 
 ciple. I have already described how you may thus obliterate 
 the object, by destroying all difference in shading. We Come 
 then to this dilemma, that if the aperture is small we cannot 
 see the object, for one reason ; and if we have the aperture 
 large we cannot see it for another reason. In one case the light 
 would break down, and in the other the object would not 
 possess any character which would enable us to- see it, and 
 thus with these very high powers we come across a state of 
 things which makes it exceedingly difficult to go much beyond 
 what we have obtained. In fact, I am much inclined to 
 believe, that as far as the size of the object is concerned, we 
 have pretty nearly got to our limit. I hope I may be wrong, 
 but I am very much afraid, that except under special condi- 
 tions, we can never see objects of as minute a size as would 
 be very desirable in studying certain characters of natural 
 history, and that our powers are limited by the constitution 
 of light itself. Perhaps something may be done to increase 
 our power by not using simple transmitted light. 
 
 This lecture would be incomplete if I were not to say 
 something about the difference between the monocular and 
 binocular microscopes. The disadvantage of monocular 
 instruments is of course that we use only one eye. We injure 
 one eye with using it too much, and the other by not using 
 it enough, and we also do not have the stereoscopic effect. 
 These disadvantages led to the desire to contrive some method 
 by means of which we could make use of both eyes. In the 
 Exhibition you may see a magnificent collection, lent by Mr. 
 Crisp, including, I believe, every kind of binocular instrument 
 that has been contrived. Some of these are so uncommon 
 
MICROSCOPES. 213 
 
 that the leading manufacturers do not know of their exist- 
 ence. Amongst these various instruments may be named 
 those constructed in the manner proposed by Nachet, 
 Wenham, Stephenson, Holmes, and Ahrens, but by far the 
 most usual are those made on the plan contrived by Mr. 
 Wenham. There is a small prism which can be pulled out or 
 pushed in, extending half over the object-glass. This prism 
 is so constructed that the light enters vertically, is 'twice 
 reflected, and passes out vertically, so that there is no dispersion, 
 and the beam remains colourless ; but by the second reflection 
 the light is sent up the oblique body of the microscope at such 
 an angle that the upper end of the two bodies are nearly at 
 the distance of our two eyes. The light which passes through 
 the other half of the object-glass not covered by the prism 
 passes up the direct body of the instrument. 
 
 One great advantage in this system of Mr. Wenham is that 
 the light which passes straight up one of the tubes is not in 
 any way influenced by the prism; and the definition is there- 
 fore unimpaired. The definition of that half of the light which 
 passes through the prism is somewhat impaired, but this does 
 not signify very much, since the vision of the left eye is mainly 
 important in enabling us to distinguish differences in the 
 level of different parts of the object, and that of the right eye 
 gives good definition of the minute detail. Another great 
 advantage of this method is that you can at once convert the 
 instrument into a monocular microscope. In some cases it 
 is very desirable only to use one tube, and this you have a 
 means of doing by pulling out the prism. This is no doubt 
 the chief reason why this form of binocular has been so much 
 more commonly employed than any other, although I am by 
 no means certain that other forms are not to some extent 
 better, at all events when high powers are used. 
 
 I have been informed that Mr. Stephenson's arrangement 
 gives exceedingly good results with very high powers, which 
 the other one will not. In it there are two truncated rectan- 
 gular prisms close behind the object-glass, so placed that they 
 reflect the light up the two bodies, both inclined to the direct 
 line of vision. By this reflection the object is inverted laterally, 
 and is still further inverted longitudinally by a larger prism, 
 so that finally the object is seen in its natural position, being 
 inverted by the object-glass and re-inverted by the prisms. 
 This instrument has the advantage of giving good results with 
 
214 LECTURES TO SCIENCE TEACHEES. 
 
 high powers ; it acts as it were like an erecting glass, and when 
 inclined at a convenient angle the stage is horizontal, which 
 is very convenient in examining objects in liquid. It is 
 thus a binocular erecting microscope, as Mr. Stephenson calls 
 it. A great many other forms have been proposed, which 
 can be seen in the Exhibition, some having one advantage, 
 and some another, whilst perhaps we may say that some have 
 none at all. 
 
 In conclusion, I may mention the application of spectrum 
 analysis with the microscope. If you want to ascertain more 
 exactly the nature of the light transmitted by any coloured 
 object, you can do so by this means. In the eye-piece arrange- 
 ments there are many things which are required only for special 
 purposes, and but for that the instrument might be made much 
 more simple. It is constructed for carrying on all kinds of 
 inquiries in this subject, and for measuring and comparing 
 spectra together side by side. You can take out the ordinary 
 eye-piece and put in the spectrum eye-piece ; or with a bino- 
 cular microscope you may see the object in its natural form 
 with one tube, and see the spectrum with the other. I have 
 for some years chiefly used my binocular arrangement, which 
 for working is undoubtedly the most convenient, but some 
 parts of its construction are rather unusual and the makers do 
 not particularly like making it, because they do not exactly 
 understand some of the necessary details. 
 
 As I said at first, the entire subject of the microscope is so 
 great that it is quite impossible in the course of an hour to do 
 more than give a hasty glance at some of the leading features. 
 In this Exhibition we have the means of studying all that we 
 can desire. We have the simplest and the most complex, the 
 earliest and the most recent of microscopes ; and I may say I 
 feel tempted to spend many hours in going into the details, in 
 order to thoroughly understand the merits and demerits of these 
 magnificent instruments. I hope that what I have said may 
 enable you to better understand their construction and be of 
 some practical use. I am quite sure that, in the present state 
 of our subject, what we want is to know something of the 
 general principles of the construction of instruments ; but still 
 more to understand the very great importance of proper illumi- 
 nation and of using object-glasses of a proper kind. I must 
 also say, that it is most important to regulate the power which 
 you use for particular purposes. It is mere child's rlav to trv 
 
MICROSCOPES. 
 
 215 
 
 to make a thing look bigger than it is, and to say you have 
 magnified it so many times. What we want is to use the 
 magnifying power that will show to the greatest advantage 
 each particular structure which we want to examine. You 
 should use the lowest power that will enable you to do that, 
 since it is often very important to see as much as possible of 
 the object, so as to understand the bearings of one part on the 
 other, and for this you must use low powers ; but when 
 you want to define certain very minute structures, you must 
 use high powers. You should always try first a low power 
 and see what that shows, and then examine with a higher 
 power in order to resolve structures that cannot be resolved with 
 lower powers ; but never use high powers unless there is some 
 evidence of such minute structure as cannot be properly seen 
 with low. 
 
ELECTROMETERS. 
 
 TWO LECTURES. 
 
 BY J. T. BOTTOMLEY, M.A., F.R S.E., DEMONSTRATOR OF NATURAL 
 PHILOSOPHY IN THE UNIVERSITY OF GLASGOW. 
 
 LECTURE I. 
 
 WHEN a difference of electric potentials exists between 
 two points it gives rise, or it may give rise, to one of two 
 effects : an electric current may be produced, or there 
 may be an exhibition of electrostatic force. Either of 
 these effects may be employed for measuring differences of 
 electric potential, and hence we have two classes of instru- 
 ments for this purpose. Thus we have galvanometers, 
 electro-dynamometers, voltameters, in which the difference 
 of potentials between two points is inferred from the 
 measurement of the current passing between those points 
 under known circumstances. On the other hand, we have 
 electroscopes and electrometers which indicate and measure 
 differences of electric potential by means of the effects of 
 electrostatic force ; that is by means of the attractions or re- 
 pulsions observed between electrified bodies. An electroscope 
 merely gives an indication of the existence of the difference 
 of potentials ; an electrometer, properly so called, measures 
 the amount of the difference. The first requisite of an 
 electrometer is to furnish numbers, by the scale readings, 
 which are proportional to the difference of potentials 
 between the points tested. From these numbers we must 
 
ELECTROMETERS. 217 
 
 afterwards deduce numbers giving differences of electric 
 potential in absolute measure, in accordance with the 
 system of absolute measurement now adopted in every 
 branch of physical science. This I will endeavour to explain 
 more fully a little later. 
 
 The first electrometric measurements were made by 
 Coulomb with the celebrated torsion balance. I have here 
 an instrument similar to that which he used. 1 
 
 The torsion balance was first devised by Mitchell, who 
 used it for the measurement of the force of gravitation 
 between two small bodies. Cavendish also employed it 
 for the same purpose. Coulomb, however, independently 
 reinvented the torsion balance, and, as a preliminary, in- 
 vestigated with great care the laws of torsional elasticity 
 for the suspending fibre. He then used the balance for 
 determining the laws of electric attraction and repulsion ; 
 and afterwards employed it for purely electrometric 
 purposes. 
 
 From the " torsion head " at the top of this tall glass 
 tube which surmounts the main body of the instrument 
 there hangs a vertical wire or glass fibre. Coulomb used a 
 fine silver wire. A fine glass fibre, as used by Faraday, was 
 first proposed by Ritchie, and is always employed now. To 
 the lower extremity of the fibre is attached a very light 
 horizontal bar or lever of shellac, which carries at the 
 extremity of the longer arm of the lever a small pith-ball 
 well gilded, and at the other extremity a counterpoise ; 
 and, in the instruments of Coulomb and Faraday, a vertical 
 disc of paper, or a slip of tissue paper hanging down ver- 
 tically, to " damp out " vibrations and bring the torsion 
 arm quickly to rest after disturbance. The case of the 
 instrument is, as you see, a glass cylinder covered with a 
 circular glass plate, into which is cemented the glass tube 
 that carries the torsion head. Round the glass cylinder 
 and on a level with the torsion arm there is pasted a 
 scale divided into degrees. By means of this scale the 
 position of the torsion arm is read off. 
 
 In the circular glass cover of the cylindrical case a 
 circular hole is cut It is for the purpose of introducing 
 what I shall call the " carrier ball." The carrier ball is a 
 
 1 The diagram represents Coulomb's original instrument. 
 
218 
 
 LECTURES TO SCIENCE TE AGREES. 
 
 little pith ball as nearly as possible of the same size as the 
 ball at the end of the torsion arm, and also carefully gilded. 
 It is fixed to the extremity of a fine rod or stem of shel- 
 lac, which is of such a length as to bring the centre of the 
 carrier ball, when in position, precisely to the level of the 
 centre of the torsion ball ; and there is a proper geometrical 
 arrangement for placing the carrier ball again and again 
 
 FIG. 1. 
 
 in precisely the same position, with its centre, let us say, at 
 the same distance from the centre of the circle in which 
 the torsion ball moves as is the centre of the torsion ball, 
 and with one side in contact with the torsion ball when 
 the middle of the torsion ball is opposite to the zero of the 
 scale upon the cylindrical glass case. Lastly, as to the 
 torsion head. The cylindrical piece, H, is cemented into the 
 
ELECTROMETERS. 219 
 
 vertical glass tube at the top. The disc, M M', graduated at 
 the circumference, is supported on the tube N, which fits 
 into the cylinder H. K is a button which fits into the hole 
 in the disc M M'. The glass fibre which carries the torsion 
 arm is attached to the button K ; and there is also an index 
 i, which indicates on the graduated scale of MM' the angle 
 through which the button K has been turned with respect 
 to it. 
 
 In Coulomb's ordinary method of using the instrument, 
 the torsion ball is completely diselectrified. The index i is 
 put at zero on the torsion head. The tube N, carrying, as 
 I have explained, the scale, button, and fibre, is turned 
 round so that the middle of the torsion ball shall be at zero 
 on the lower scale. The carrier ball is now electrified and 
 put into its place, and the torsion ball comes in contact 
 with it. The charge is thus divided between the two balls 
 halved if the balls are exactly equal and similar and the 
 torsion ball is repelled from the other. The displacement 
 of the torsion arm is opposed by the torsion of the fibre, 
 and finally we have equilibrium between the couple, as 
 it is called in dynamics, due to the repulsion of the balls 
 acting at the end of the torsion arm, and the couple due 
 to torsion. By turning the button K of the torsion 
 head, the torsional couple can be increased, which will 
 have the effect of forcing the torsion ball nearer to the 
 carrier ball. 
 
 Coulomb, as I have said, first undertook the determination 
 of the laws of electric attraction and repulsion. He began 
 by examining very carefully the laws of torsional elasticity 
 of the wires that he employed, and established the im- 
 portant law that the torsional couple, for a given wire, is 
 in simple proportion to the angle of torsion ; or, in other 
 words, that if a certain couple be required to turn the lower 
 end of the wire through a certain angle relatively to the 
 upper end, the couple required to turn it through twice that 
 angle will be double ; the couple required to turn it through 
 three times that angle will be triple of the first, and so on. 
 He tested this law to high angular displacements, and found 
 that it holds with great exactness unless the torsion is so 
 great that the wire receives a permanent set, and when 
 released from torsion does not return to its original 
 condition. 
 
220 LECTURES TO SCIENCE TEACHERS. 
 
 The laws of torsion being established, the first experi- 
 ments of Coulomb were on the force of electric repulsion at 
 different distances. The balls being electrified as I have 
 explained, the torsion ball takes up a position in which the 
 repulsion couple is balanced by the couple of torsion. The 
 distance between the balls was determined from reading off 
 the position of the torsion ball by means of the scale round 
 the glass cylinder. By a simple trigonometrical calculation 
 the distance is deduced from this reading. The torsion 
 head was then turned, so as to reduce the distance, which 
 was again measured in the same way, and the torsion couple 
 at this new distance was again determined. The torsion 
 couple required to equilibrate the repulsion between the 
 balls at different distances, the charge being supposed to 
 remain unaltered, was thus obtained. By these experi- 
 ments, the results of which, however, required correction 
 for the inevitable loss of charge during the experiment, 
 Coulomb obtained his well-known law of the inverse square 
 of the distance. 
 
 This law, I must remark, requires careful exactness in 
 its statement. Through a complete misunderstanding of it 
 Sir W. Snow Harris was led to experiments to disprove 
 it. A proper statement is, that if the two quantities of elec- 
 tricity are placed upon balls which are so small that the 
 diameters of the balls are insensible in comparison with 
 the distance between their centres, the force of repulsion is 
 inversely proportional to the square of the distance between 
 the centres of the balls. Even this statement must be 
 taken with limitations, because the effects of induction 
 prevent the experiments from showing precisely that law. 
 There are, however, other proofs, partly given by Coulomb, 
 partly by Cavendish, which establish the truth of the law 
 with minute exactness. 
 
 The mathematical theory for a particular form of 
 torsion balance, taking induction into account, has been 
 given by Professor Clerk-Maxwell in his great work on 
 electricity and magnetism. 
 
 The next part of Coulomb's investigation consisted in 
 determining the effect of electrifying the balls with different 
 quantities of electricity, and to do that he used an ingenious 
 device. Electrifying the carrier ball, and bringing it to 
 the electrometer, the movable or torsion ball received its 
 
ELECTEOMETEES. 221 
 
 charge. He now removed the carrier ball, and touched it 
 with an insulated ball exactly equal to it in dimensions ; 
 he was thus able to halve the charge of the carrier. Putting 
 it back, he tested the force. Again the charge of the 
 carrier was halved, and again he determined the force, and 
 so on. The charge on the torsion ball was then altered, 
 and a new series of experiments proceeded with. By a 
 series of experiments of this kind he found that the force 
 between the two balls is proportional to the product of 
 the quantities of electricity on them, the distance remaining 
 the same ; and combining the latter with the former law, 
 he showed that the force is proportional to 
 
 m x m! 
 
 d 2 
 
 where m and m' are the charges of the balls, and d the 
 distance between their centres. But in stating this law you 
 must be careful to mention the limitations as to the 
 dimensions of the balls and as to effects of induction to 
 which I have just referred. 
 
 The experiments that I have been speaking of were not 
 so much electrometric experiments as experiments on the 
 fundamental laws of electricity ; but the next experiments 
 of Coulomb were of a purely electrometric character. They 
 were for the purpose of determining the distribution of 
 electricity over variously shaped conductors. Except in cer- 
 tain simple cases, the mathematical problem of determining 
 the distribution of electricity over a conductor is extremely 
 difficult. Poisson had, however, worked out the case for 
 two equal spheres in contact, assuming the truth of 
 Coulomb's law of the inverse square of the distance. Now 
 it is plain that nothing could give a better test of the truth 
 of the law than the comparison of the distribution given 
 by the calculation of Poisson with the distribution deter- 
 mined by experiment. 
 
 Coulomb used for these experiments another instrument, 
 slightly different from the carrier ball, namely, the proof- 
 plane, which consists simply of a very small piece of thin 
 metal insulated on a thin arm of shellac, which is applied 
 to the surface of the conductor to be tested, and then 
 carried away from the conductor. It is then brought to 
 
222 LECTURES TO SCIENCE TEACHERS. 
 
 the torsion balance and placed in position, and the amount 
 of electricity that it possesses is determined. The theory 
 of the proof -plane shows 'that when the proof -plane is applied 
 to the surface of the body to be tested, it, as it were, 
 carries away a small portion of the surface of the body 
 with its electricity upon it. While the proof -plane is in 
 contact with the body it forms a part of the surface ; and 
 when it is lifted normally from the surface of the body with 
 which it is in contact, it is practically a small portion 
 of the surface of the body itself that is carried away. 
 Coulomb, by an error, supposed that the proof -plane carried 
 away as much electricity on each side as corresponds to the 
 space it touched upon. To correct his results for this error 
 is, however, easy. 
 
 Taking, then, the case of the two spheres in contact 
 worked out by Poisson, and the case of an ellipsoid, for 
 which Coulomb himself worked out the distribution mathe- 
 matically, he found, on determining the distribution by 
 experiment, as close an agreement with the theoretical 
 distribution as could possibly be expected. 
 
 By far the best proof that we have of the exactness of 
 the laws of electric attraction which I have just stated is an 
 indirect proof. It was first pointed out by Cavendish that 
 if the law of the inverse square of the distance be true it 
 follows as a consequence that the whole of the electricity 
 upon an electrified closed conductor must reside at the surface 
 of the conductor. The exactness of the law will then be 
 tested by finding whether or not the whole of the electricity 
 does reside at the surface of such a conductor. Cavendish 
 himself undertook some admirable experiments to test this 
 question, and some of the most important of Faraday's 
 electrostatic researches were devoted to it. 
 
 One of the remarkable experiments of Faraday was to 
 place himself within a chamber a twelve-foot cube 
 constructed of light material and covered with tinfoil, 
 taking with him electroscopes of the most delicate con- 
 struction to examine whether any electrification could be 
 found in the interior while the chamber, which was 
 insulated, was electrified ; but while the whole power of the 
 electric machines of the Royal Institution was turned on, so 
 that from every part of the outside of the chamber flashes 
 and brushes of electricity were rushing off, he could find 
 
ELECTROMETERS. 223 
 
 no trace whatever of electrification on the walls or else- 
 where within. The most refined modern experiments 
 lead to the same result ; and it is evidence of this kind that 
 enables us to regard the law given by Coulomb as holding 
 with extreme exactness. 
 
 In connection with experiments of this kind Faraday 
 used the torsion balance and made great improvements 
 in it. He used the torsion balance in those great 
 researches in which he investigated the curved lines of 
 force, and worked out the theory of induction, and proved 
 that electrification by induction is not what it was sup- 
 posed to be action at a distance but that it is electric 
 disturbance transmitted by means of contiguous particles 
 of the di electric, or insulating medium, between two electri- 
 fied bodies One of the modifications that Faraday made 
 in the torsion balance was the protection of the movable 
 parts of the instrument from external influence. He 
 showed that the indications of an instrument such as the 
 torsion balance, or common gold-leaf electroscope, are per- 
 fectly untrustworthy unless the most careful attention is 
 paid to this matter ; and it is curious that though it is now 
 forty-five years since Faraday pointed this out, instrument- 
 makers have failed, except in very few cases, to give any 
 attention to his warning. For instance, here are two 
 well constructed instruments a Peltier's electrometer and 
 a torsion balance, and neither of these has the slightest 
 protection. Out of all the instruments for teaching pur- 
 poses that we see in this magnificent Loan Collection, 
 electroscopes, electrometers, and so forth, you will find 
 not one in a hundred to have any pretence of protection 
 whatsoever. 
 
 I have here a very simple experiment for illustrating the 
 importance of guarding the movable parts of your electro- 
 static instruments from external influence. Here is a 
 glass bell-jar, from the roof of which hangs a fine glass 
 fibre carrying a horizontal arm of aluminium. The inside 
 of the bell-jar is very clean, and is kept dry with the aid 
 of a dish of strong sulphuric acid. It thus gives excellent 
 insulation, and the aluminium needle, which was charged 
 some hours ago, has, I have no doubt, retained its charge well. 
 I now dry the outside of the bell-jar by passing over its 
 surface a wire hoop, covered with cotton wick, and flaming 
 
224 LECTURES TO SCIENCE TEACHERS. 
 
 with spirits of wine. Now if I move this glass rod slightly 
 electrified, or this electrified stick of sealing-wax, in the 
 vicinity of the aluminium arm, you see at once the effect 
 of external electrification on the movable arm, which 
 corresponds exactly to the movable arm in the torsion 
 balance. I will now bring the glass rod near to the 
 outside of the bell-jar, and draw this slip of paper over 
 the surface of the bell- jar in the vicinity of the glass 
 rod, and then carry away the glass rod. You see that I 
 have left the bell-jar itself with a portion of its outer sur- 
 face electrified in such a way that it attracts the needle 
 round through a right angle. 
 
 I will now wet the outside of the bell-jar with a sponge 
 and thus make the outside of the glass a fairly good 
 conductor. By this the needle is partly, but as you see not 
 entirely, screened from the influence of external electri- 
 fication. Yet the natural dampness of the glass cases, 
 the dust and so forth that may be upon them, is the 
 whole protection that the electroscopes and electrometers 
 ordinarily constructed have against such disturbing 
 influence. 
 
 Faraday pointed out that nothing but a complete metallic 
 protection is sufficient. He protected the whole of the 
 interior of the electroscopes and electrometers that he used 
 with slips of tinfoil pasted on the inside surface of the 
 glass cases, leaving only such spaces as were necessary for 
 looking through. In the instruments of Sir William 
 Thomson that I shall have to show you the movable 
 parts are protected by brass covers, tinfoil, or by a wire 
 cage which is often found to be sufficient and to be very 
 convenient. 
 
 Another improvement that Faraday made with respect to 
 the torsion balance was the introduction of what is called the 
 heterostatic method of using the instrument. Electrometers 
 may be distinguished into two classes, idiostatic and hetero- 
 static. In the idiostatic class the whole electric force 
 depends on the electrification which is itself the subject of 
 the test. In the heterostatic class, besides the electrification 
 to be tested, another electrification, maintained indepen- 
 dently of it, is taken advantage of. The torsion balance as 
 used by Coulomb was an idiostatic instrument. The elec- 
 tricity to be tested was divided each time between the balls 
 
ELECTROMETERS. 225 
 
 and was itself the cause of the electric force. Faraday 
 preferred to employ the instrument thus. He gave a per- 
 manent charge to the torsion ball and placed it in a definite 
 position, suppose 30 from the zero mark on the scale pasted 
 round the cylindrical case of the instrument. He then 
 introduced the carrier ball or proof plane, bearing its 
 charge, which of course displaced the torsion ball by attrac- 
 tion or repulsion. Turning the button of the torsion head 
 the torsion ball was brought back to its first position, and 
 the torsion required in order to do this was determined. 
 
 In his celebrated Experimental Researches on Electricity, 
 Faraday describes minutely the precautions that must be 
 taken in using the torsion balance. He considered it a very 
 valuable instrument in good hands : and in his hands it 
 proved indeed a most valuable instrument. It was by 
 means of it that all his grand series of electrostatic re- 
 searches was carried on. 
 
 From the torsion balance of Coulomb there came a 
 variety of other instruments more or less like it. Here for 
 example is Peltier's electrometer, which if it had protection 
 of the movable parts would be an excellent instrument. 
 Instead of an arm supported on a torsion fibre we have here 
 a long aluminium needle pivotted on a fine needlepoint, and 
 to the movable needle there is attached a very short magnet. 
 The small magnet gives the directive force instead of the 
 torsion of the fibre. To use the instrument it is set with 
 the magnet in the magnetic meridian and the two movable 
 arms in contact with these two repulsion balls which take 
 the place of the carrier ball in the torsion balance. When 
 a charge is given to the instrument by means of this 
 charging rod connected with the repulsion balls, the charge 
 is also communicated to the movable needle, and repulsion 
 is the result. The electric repulsion is balanced by the 
 tendency of the magnet to return to its normal position in 
 the magnetic meridian. 
 
 Here again is a similar instrument by Kohlrausch. It 
 consists also of a conductor, a'portion of which is a magnet, 
 and repulsion plates similar to the two repulsion balls which 
 were used in the Peltier electrometer. Kohlrausch calls 
 this instrument a sine electrometer because in his way of 
 using it the forces are proportional to the sines of the angles 
 of deflection of the movable needle. 
 
226 LECTUEES TO SCIENCE TEACHERS. 
 
 Of repulsion electrometers I have lastly to bring before 
 you those of Sir William Thomson. The very remarkable 
 collection of electrometers exhibited in the Loan Collection 
 by Sir William Thomson shows the attention that he has 
 bestowed on electrometry. Electrometers in every stage of 
 development may be seen in his collection, but many of 
 them you will have to go and look at in their place. It 
 was impossible to bring them all upon the table before you ; 
 and much more was it out of the question for me to think 
 of attempting to explain them in the two lectures devoted 
 to this subject. 
 
 Modern electrometry is largely due to Sir William 
 Thomson. His instruments have for the present superseded 
 all other electrometers for practical purposes, such as 
 the testing of telegraph cables during construction and 
 after submersion ; and I wish to call your attention to his 
 admirable report on " Electrometers and Electrostatic 
 Measurements," prepared for the British Association Com- 
 mittee on Standards of Electrical Resistance (1867), and 
 republished in his collected papers on Electrostatics and 
 Magnetism. In that report you will find full information 
 on many points that I cannot do more than allude to, while 
 exhibiting some of the electrometers of the Loan Collection 
 to you. 
 
 We have three specimens of Thomson's repulsion electro- 
 meters before us. Two of them are almost identical in 
 construction, so I have dissected one that you may see its 
 parts. 
 
 The glass case is a thin flint glass bell, the lower half of 
 which is coated inside and outside with tinfoil, like a Leyden 
 jar, except that at the bottom of the inside a part of the 
 glass is left bare. That part is filled with strong sulphuric 
 acid, and connected by a piece of platinum foil with the 
 tinfoil coating. The sulphuric acid performs two functions. 
 It keeps the inside of the case of the instrument dry, pre- 
 venting the deposit of moisture on the glass insulators, 1 
 
 1 It is curious that in almost all even of the most recent text-books 
 on electricity we still find a distinction made between diy air and damp 
 air as insulators. So far as we know at present, no difference what- 
 ever exists. Sir William Thomson has not been able to detect the 
 slightest difference between dry air and damp air as to power of insu- 
 lation by the most delicate experiments. The cause of the widespread 
 fallacy on this subject is, of course, that such conductors as are used 
 
ELECTROMETERS. 
 
 227 
 
 and it also acts as a part of the inside coating of the 
 Leyden jar, like the water in the celebrated experiment 
 that led to the discovery of the Leyden jar. 
 
 The glass bell is enclosed in a metal case which supports 
 it and protects it, and which is furnished with three level- 
 ling screws as feet of the instrument. The case is covered 
 in with a circular plate of glass in a metal rim. At the 
 centre of this glass plate is supported a torsion head, as in 
 Coulomb's torsion balance, having an index protruding out 
 
 FIG. 2. 
 
 so as to move over graduations on the circular metallic 
 ring. 1 
 
 From the torsion head hangs a fine glass fibre, to which 
 is attached the movable part of the electrometer. This 
 consists of a light horizontal needle made of aluminium. 
 To it there is attached a stiff platinum wire, hanging down 
 
 for electrical experiments are found to maintain a charge much better 
 in dry than in damp air. The extraordinary loss observed in damp 
 air is, however, altogether by means of the supports (of glass, shellac, 
 &c.), the surfaces of which receive a conducting film, often invisible, 
 of moisture from the air. 
 
 1 The parts of the instrument were shown in the lecture. They will 
 be easily distinguished in the accompanying diagram. 
 
 Q 2 
 
228 LECTURES TO SCIENCE TEACHERS. 
 
 vertically and nearly reaching the bottom of the jar, and a 
 fine platinum wire with a little weight of platinum attached 
 to it hangs down from the end of the stiff wire and dips 
 into the sulphuric acid at the bottom. Thus the needle is 
 kept connected with the inside coating of the Leyden jar. 
 
 Bather less than half-way from the bottom of the jar 
 there is a metallic ring, cemented to the inside for the pur- 
 pose of supporting two repulsion plates. This ring and the 
 repulsion plates are connected with the interior tinfoil 
 coating. When, therefore, the jar is charged with electricity, 
 repulsion takes place between the needle, also connected 
 with the inside coating, and the repulsion plates, and the 
 needle is driven up against two stops, connected with the 
 plates, which limit its motion. 
 
 The ends of the needle and the repulsion plates are sur- 
 rounded by a cage of fine brass wire, which is stretched on 
 a brass framework. The framework is supported from the 
 metallic case of the electrometer by two glass pillars. It 
 is insulated from the case and from the needle and repulsion 
 plates ; and it has an electrode projecting outward through 
 a hole in the outer case, by means of which the cage can 
 be connected to any body that is to be tested. 
 
 Let me now explain the use of the instrument. The 
 Leyden jar is charged by means of an electrode provided 
 for the purpose, and, as I have said, the needle is repelled 
 from the repulsion plates against the stops. The electrode 
 of the cage is connected with the earth, or more usually 
 with the brass case of the instrument, which is connected 
 with the earth. The torsion head is now turned in such a 
 way as to oppose the torsion of the wire to the electric 
 repulsion, and the needle is forced away from the stops. 
 The observer, looking down through the plate glass cover, 
 now brings the needle, by means of two marks, one on the 
 glass cover and the other on a ring below, into a " marked 
 position," and then reads off the number of degrees of 
 torsion on the scale over which the torsion index moves. 
 This reading is commonly called the " earth reading," as 
 the tested conductor the cage is in connection with the 
 earth during the observation. The number of degrees of 
 torsion required to bring the needle into the marked position 
 depends upon the electrification of the jar ; and it can readily 
 be shown that the potential to which the jar is charged, 
 
ELECTROMETERS. 229 
 
 that is the difference of potentials between the jaT and the 
 cage or the earth, is measured by the square root of the 
 number of degrees of torsion. When I say that the 
 potential of the jar is measured by the square root of the 
 number of degrees of torsion, I mean that if the jar be 
 electrified to different potentials the square roots of the 
 torsional readings will be in proportion to those potentials. 
 
 Now, the charge of the jar being kept unaltered, the 
 electrode of the cage is disconnected from the earth and 
 connected with the conductor to be tested, and the cage 
 thus acquires the same potential as the conductor to be 
 tested. The electrification of the cage causes an alter- 
 ation in the force acting upon the needle which moves 
 towards or from the repulsion plates, according as the 
 electrification of the cage is similar or dissimilar to that 
 of the jar. The torsion head is now turned so as to 
 bring the needle back to the sighted position ; and the 
 torsion is again read off. As before, the square root of 
 the number of degrees of torsion measures the difference 
 of potentials between the cage, and the needle, that is also 
 the difference of potentials between the conductor tested 
 and the needle. The excess, positive or negative, of the 
 square root of this last reading above the square root 
 of the "earth reading," is the difference of potentials 
 between the body tested and the earth ; and if, as is com- 
 monly done, we regard the potential of the earth as zero, 
 it is simply the potential of the conductor tested. 
 
 This electrometer has done good service in observations 
 on atmospheric electricity. Indeed it was first constructed 
 for that purpose. The electrode of the cage is connected 
 with a collector of electricity, either with a flame-collector, 
 as first described by Volta, or with the water dropping 
 collector of Sir William Thomson, and earth readings and 
 atmospheric readings are taken alternately at proper inter- 
 vals. The numbers obtained for the differences of potentials 
 are reduced to absolute measure by comparison with a few 
 cells of a known battery; but this I will explain a little 
 later. 
 
 [A form of portable electrometer, also due to Thomson, 
 was next exhibited and described. The description is not, 
 however, reproduced here, as the instrument differs little 
 in electrical principles from that which has just been 
 
230 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 spoken of. The suspension of the needle is different. The 
 needle is firmly attached at right angles to a tightly- 
 stretched platinum wire, the lower end of which is fixed, 
 and the upper mechanically connected with, though elec- 
 trically insulated from, the parts that correspond to a 
 torsion head. See Thomson's Electrostatics and Magnetism, 
 xvi., 277.] 
 
 I will conclude my lecture for to-day by showing you an 
 electrometer, the first of a new class, and one of great 
 interest. This is Thomson's first " divided ring " electro- 
 meter, 
 
 Fio. 3. 
 
 In this instrument there is a broad metallic ring cut 
 into two parts, and each half ring is supported horizontally 
 on two vertical pillars of thin glass rod. One half ring is 
 kept connected with the case of the electrometer ; the other 
 can be connected by means of a proper electrode with any 
 body to be tested. From the top of the case of the electro- 
 meter hangs a fine glass fibre which carries a light 
 aluminium needle, projecting from a point a little above 
 the centre of the divided ring. The needle being counter- 
 poised projects out on one side only. It is sufficiently long 
 to project out over the divided ring. 
 
ELECTROMETERS. 231 
 
 A stiff piece of platinum wire attached to the needle at 
 right angles to it hangs or projects vertically down ; and from 
 the lower extremity of this a very fine piece of platinum 
 wire hangs down and dips into the strong sulphuric acid which 
 forms the interior coating of a Leyden jar. The case of 
 the instrument is of glass, with slips of tinfoil pasted over 
 it to protect the interior from external influence. There are 
 other particulars to which I need not allude just now, as 
 I shall have to speak of them in connection with the 
 quadrant electrometer, a development from that which is 
 before us at present. 
 
 Before the Leyden jar with which the needle is connected 
 is electrified, the needle is adjusted mechanically, so that 
 it projects over one of the divisions between the two halves 
 of the divided ring. Now if the two halves of the ring 
 are at the same potential, both being, let us suppose, con- 
 nected with the case of the instrument, and if the parts of 
 the instrument are quite symmetrical, then when the Leyden 
 jar and needle are electrified, no disturbance of the needle 
 will be experienced. But let one of the half -rings be con- 
 nected with a body to be tested whose potential differs 
 from that of the case of the instrument to which the other 
 half-ring is connected, and there will plainly be attraction 
 or repulsion of the needle, according to the nature of the 
 electrification of the tested body, and the needle will move 
 towards one half-ring or towards the other. 
 
 Now let me in conclusion refer to one very remarkable 
 application that Sir William Thomson has made of this 
 electrometer. It was chiefly for that purpose that I have 
 brought it before you, because the instrument itself, though 
 in earlier times it did admirable service as an electrometer, 
 has now been superseded by the quadrant electrometer. 
 By means of this instrument Sir "William Thomson was 
 able to furnish a test between Volta's contact theory of his 
 pile and the rival chemical theory. The fundamental 
 statement of Volta's theory was that a piece of metallic 
 zinc and a piece of metallic copper put in contact assume 
 different potentials, the zinc becoming positive with respect 
 to the copper. The supporters of the chemical theory 
 denied this, and explained away the experiments that were 
 adduced in support of it. It is now, however, regarded as 
 established, and here is one of the experiments of Thomson 
 
232 LECTURES TO SCIENCE TEACHERS. 
 
 to prove its truth. Taking away the two half-rings of 
 brass he substituted this ring which I now show you, one 
 half of which is of copper and the other half zinc, the two 
 soldered into one ring. The needle and Ley den jar of the 
 electrometer being completely discharged, the needle was 
 mechanically adjusted so as to hang with its extremity over 
 one of the soldered joints between zinc and copper. The 
 needle and jar were then electrified, first positively and then 
 negatively, and it was found that, when positive the needle 
 moved from the zinc half-ring towards the copper half-ring, 
 and when electrified negatively it moved from the copper 
 half-ring towards the zinc half-ring. Thus the zinc arid 
 copper half-rings in contact behaved as would two insulated 
 half-rings of the same metal, one of which, corresponding 
 to the zinc, is electrified positively relatively to the other. 
 
ELECTROMETERS. 
 
 LECTURE II. 
 
 THE instrument to which I will now direct your attention 
 is Thomson's quadrant electrometer. It is the most recent 
 and most complete development from the divided ring 
 electrometer which we examined yesterday. The semicircles 
 of the divided ring over which the movable needle swings 
 are in the quadrant electrometer replaced by four quadrants 
 of a hollow cylindrical box, within which is the movable 
 needle ; and the needle, instead of projecting on one side 
 only of the axis of suspension, as in the divided ring elec- 
 trometer, is symmetrical about that axis. 
 
 In the quadrant electrometer we have first a white flint 
 glass bell- jar, surrounded and supported, mouth up, by a 
 metal casing. The outside is partially coated with tin-foil ; 
 the inside contains strong sulphuric acid a couple of inches 
 deep. This arrangement gives us, as we have already 
 seen in former cases, a Leyden jar ; and with the in- 
 side coating of the Leyden jar the movable needle is 
 connected. 
 
 Secondly, we have the " main cover " of the instrument, 
 a circular brass plate, covering the mouth of the jar and 
 screwed down to the metal casing of the jar. Over a large 
 circular hole in the main cover stands what is called the 
 " lantern." The lantern is of brass, but with a window in 
 front. It carries the " gauge," and three electrodes project 
 from the top of it. [See Fig. 4.] 
 
 The four quadrants (Fig. 5) are supported each on a short 
 pillar of glass projecting downwards from the main cover of 
 the electrometer. The supports of these glass pillars are 
 movable in radial slots, and the quadrants can be drawn 
 out from or pushed in towards the axis of suspension. 
 When the instrument is in adjustment they are pushed in 
 
234 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 as near as possible without touching, and thus form parts of 
 a cylinder, divided only from each other as it were by two 
 saw-cuts. Thus also they are arranged symmetrically 
 around the needle. One of the four is capable of adjust- 
 ment by a screw turned by a milled bead. Slight motions 
 of this quadrant serve for adjusting the zero of the instru- 
 ment occasionally. 
 
 The quadrants are connected in pairs by fine wires, the 
 opposite quadrants being connected together ; and from one 
 
 Fio. 4. 
 
 quadrant of each pair rises an electrode which passes insn- 
 tated through the top of the lantern. These electrodes serve 
 to connect two conductors, whose difference of potentials 
 is to be measured, one with each pair of quadrants. The 
 third electrode (see Fig. 4), passing up through the top 
 of the lantern, is for charging and discharging the Leyden jar 
 of the electrometer. 
 
 We next come to the " needle." It is a thin flat piece 
 of sheet aluminium, shaped perhaps more like the paddle 
 
ELECTROMETERS. 
 
 235 
 
 of a canoe than like a needle (Fig. 5) ; but, you know, in 
 electricity and magnetism the name ( needle is commonly 
 applied to bodies that in shape and size are very unlike 
 that from which the name is taken. The needle is borne 
 horizontally on a stiff vertical platinum wire, which passes 
 through its centre upward and downward. The stiff platinum 
 wire is attached to a small cross-bar, which is carried by a 
 bifilar suspension. [See Fig. 6, where the upper end of the 
 stiff platinum wire is seen coming up through the " guard 
 tube " and having the " mirror " attached. The bifilar 
 fibres of the suspension are seen coming up to two pins, 
 c and d.] To the lower end of the stiff platinum wire a very 
 fine platinum wire is attached. This carries a small plati- 
 num weight, which hangs down and dips into the sulphuric 
 
 Fio. 5. 
 
 acid at the bottom of the jar; and the needle is by this 
 means kept in connection with the interior coating of the 
 charged Leyden jar. 
 
 I have got here the suspension plate of an electrometer 
 to show you (Fig. 6). It is supported and insulated by 
 a glass pillar, which rises from the main cover, and it is 
 covered by the lantern. On the face of it are five pins, 
 (a) (6) (c) (d) and (h), which can be turned by means of a 
 square pointed key. The threads of the bifilar are wound up 
 on the pins (c) (d), and (c) and (d) are capable of being turned 
 so as to adjust the lengths of the fibres to be equal. The 
 pins (c) and (d) are carried on springs (e) and (/), which are 
 screwed to the face of the suspension plate by screws shown 
 about one -third of way from the bottom of that plate. The 
 pin marked (h) is a conical plug, which passes in between 
 
236 LECTURES TO SCIENCE TEACHERS. 
 
 the springs and screws into the plate behind, and by turn- 
 ing it in or out the conical plug presses the springs (e) and 
 (/) apart, or allows them to approach. Thus the upper 
 points of the bifilar suspension are separated or brought 
 nearer together, and the sensibility of the instrument is 
 diminished or increased. The pins (a) (b) are screwed into 
 the springs (e) (/), and press against the plate behind. 
 
 FIG. 6. 
 
 When one or other of them is turned it brings forward 
 the neighbouring point of suspension, or allows it to recede. 
 Lastly, as to this part, the platinum wire carrying the 
 needle has a small circular mirror attached to it (see Fig. 
 6), and all the movable parts are carefully guarded from 
 external influence ; the platinum wire, both above and 
 below the needle, by a metal " guard tube " through which 
 it passes, and the mirror by a little cylindrical hood project- 
 ing some distance from the suspension plate. 
 
ELECTEOMETEES. 237 
 
 The mirror is a small concave mirror of silvered glass. 
 It is about the size of a threepenny-piece. It is extremely 
 light, weighing only about ,a third of a grain. To make 
 these little glass concave mirrors (which are also used for 
 reflecting galvanometers, where extreme lightness is even 
 more essential than in the present case) was a matter of 
 considerable difficulty. The following plan is now adopted. 
 A large number of little circles of the finest micro- 
 scope glass are silvered. When they are to be used for 
 reflecting galvanometers, four minute magnets are cemented 
 to the back. The mirrors are then tested, to see that each 
 gives a good image of a lamp reflected from it ; and, out of 
 perhaps fifty tried, ten or fifteen may be found satisfactory. 
 This plan of selection by trial gives mirrors that afford a 
 perfect image, and which are lighter than any that can be 
 obtained by the plan of grinding at first adopted. The 
 concavity of the mirrors is such that the rays of a lamp 
 placed about one metre (39-J- inches) from the mirror, and 
 reflected from it, are brought to a focus on a screen at the 
 same distance from the mirror. 
 
 In the use of the electrometer a lamp is placed in front 
 of the mirror, and its rays, passing through a narrow vertical 
 slit, fall on the mirror and are reflected from it. The reflected 
 rays are brought to a focus on a long horizontal screen at 
 right angles to the line from the slit to the mirror, and just 
 above the slit. The screen, which has marked on it a 
 finely-divided scale, is one metre distant from the mirror. 
 
 Now you will see that the position of the image of the 
 slit on the scale depends upon the angle at which the rays 
 proceeding from the slit to the mirror fall upon the mirror ; 
 and as the mirror turns with the needle, the position of the 
 image on the scale depends upon the position of the needle. 
 When the needle is at its position of zero deflection, the 
 image of the slit is seen at a point on the scale just above 
 the slit. When the needle is deflected to one side or other, 
 the image on the scale is also deflected, and by a well-known 
 optical principle the angle of deflection of the reflected rays 
 is twice the angle of deflection of the mirror, that is of 
 the needle. 
 
 We are now prepared to understand the principle on which 
 the use of the quadrant electrometer depends. We have 
 in this case three bodies, two of them fixed (the two pairs 
 
238 LECTURES TO SCIENCE TEACHERS. 
 
 of quadrants), and one (the needle) movable about an axis, 
 the arrangements being of the symmetrical kind that I 
 have described, and the movable body is maintained at a 
 constant high potential. The potentials of the two fixed 
 conductors are different, and it is this difference which we 
 wish to measure. Let Y denote the potential of the needle, 
 let Yj and Y 2 be the potentials of the two fixed conductors ; 
 let be the angle of deflection of the needle from the posi- 
 tion which it would occupy were V x and V 2 equal, then it 
 can be shown that if Y x be greater than Y 2 , 
 
 = 0(V 1 -.V 3 )[V -J(V 1 + V 9 )] . . (A) 
 
 where C is a constant. Now if Y be very great in com- 
 parison with Y x and Y 2 , the second term, ^ (Yj + Y 2 ), of the 
 last factor may be omitted in this expression, and we obtain 
 
 In the use of the quadrant electrometer this is practically 
 the state of affairs ; for, as I have just said, Y is the 
 potential of the Leyden jar with which the needle of the 
 electrometer is connected, and the potential of the jar is 
 always maintained very great indeed in comparison with 
 that of any electrification that this delicate instrument is 
 employed to measure. 
 
 Taking now the formula (B) we see in the first place 
 that 6, the angle of deflection, is in simple proportion to the 
 difference of potentials Y x - Y 2 of the quadrants, that is, of 
 the two bodies tested, Y being kept constant. Secondly, 
 we see that as is proportional to Y , V 1 Y 2 being 
 constant, the deflection for any given difference of poten- 
 tials is greater the greater Y . Thus the sensibility of 
 the instrument is proportional to the potential of the 
 needle. By altering the charge of the Leyden jar, therefore, 
 we may alter the sensibility of the electrometer. 
 
 To keep the charge of the Leyden jar constant, which 
 you see is essential in carrying out any set of measure- 
 ments which are to be immediately comparable, there is 
 a " gauge " connected with the electrometer, as well as 
 a minute electric machine within the glass jar, and supported 
 from the main cover, by means of which the charge can be 
 increased or diminished. The gauge is really an electro- 
 meter, precisely similar in construction to one that I shall 
 
ELEGTEOMEIERS. 239 
 
 have to describe immediately, the portable electrometer, 
 but turned upside down, and used idiostatically. Above 
 this horizontal plate (the circular plate at the top of Fig. 
 6), connected with the suspension plate of the needle, there 
 is an arrangement resembling somewhat a cart-weighing 
 machine. G is a brass plate, with a square hole cut in it. 
 The hole is over the middle of the circular plate, connected 
 with the suspension plate, p is a square of light sheet alumi- 
 nium ; it is connected with the long arm h, the square p 
 and the long arm h being in one piece. This lever is 
 carried on a tightly-stretched platinum wire /, round 
 which as a fulcrum it is perfectly free to move ; and tor- 
 sion is given to the wire/, so as to tend to raise p upwards 
 and to depress the arm h. But when the little square p is 
 attracted by the plate connected with the needle and jar, the 
 
 FIQ. 7. 
 
 attraction draws p downward against the torsion of the 
 wire / to a fixed position. 
 
 The extremity of h is cut out into a little fork, across 
 which a fine black hair is stretched, and behind the hair 
 there stands a minute white porcelain plate with two round 
 black dots upon it. The arrangement is looked at by the 
 magnifying lens I, and the plate p is in the proper posi- 
 tion when the hair is midway between the two dots. The 
 Leyden jar of the electrometer is then charged up until 
 this is the case. 
 
 The little electric machine for increasing or diminishing 
 the charge of the jar is one of a kind first invented by 
 Nicholson, known by the name of Nicholson's Revolving 
 Doubler, and used by Yolta in some of his researches. 
 Recently several such machines have been constructed. 
 That of Holtz ; now well known, makes use of the " com- 
 
240 LECTURES TO SCIENCE TEACHERS. 
 
 pound interest principle," involved in the action of the 
 Doubler. 
 
 I have just shown you that the deflections observed by 
 the quadrant electrometer are in simple proportion to 
 differences of potentials between the quadrants, or differ- 
 ences of potentials between any two conductors tested. 
 From these scale readings we must deduce numbers giving 
 the differences of potentials in absolute measure. One way 
 of doing this is to compare the indications of the quadrant 
 electrometer with those of an absolute electrometer, such as 
 I am about to describe immediately, and to deduce " the 
 constant " for the quadrant electrometer. The scale read- 
 ings may then be reduced to absolute measure by multiplying 
 by the constant ; and the constant of the instrument is the 
 same so long as the gauge is kept in the same condition, and 
 the quadrants in the same position as when the comparison 
 with the absolute electrometer was made. Another method, 
 which can at any time be applied with the greatest ease, is to 
 find the deflection given by a galvanic cell of known electro- 
 motive force, and deduce the constant in that way.- For 
 example, let a Daniell's cell be applied to the quadrant electro- 
 meter, and let us suppose that the deflection is seventy-five 
 S3ale divisions on one side of the middle position, and when 
 the electrodes of the cell applied to the electrodes of the 
 electrometer are reversed, a deflection of seventy-five divi- 
 sions on the other side of the middle is observed. Now the 
 electromotive force of a Daniell's cell, or the difference of 
 potentials producible by a Daniell's cell, is well known from 
 the experiments of Sir William Thomson. 1 It is 1'12 of the 
 unit termed by practical electricians a volt.' 2 ' Hence, divid- 
 ing yly^, we get 67 as the deflection produced on the 
 electrometer by a difference of potentials of one volt. 
 A double deflection would be produced by a difference of 
 potentials equal to two volts, and so on. 
 
 [Some of the applications of the quadrant electrometer 
 were next very briefly referred to, particularly its application 
 to observation of atmospheric electricity.] For this purpose 
 it is admirably suited. The mirror electrometer is the 
 
 1 Proc. Roy. Soc. 1860, and Reprinted Papers, xviii. 
 
 2 One volt is equal to 10 8 centimetre-gramme-second electro-mag- 
 netic units. See Everett's Illustrations of the Centimetre Gramme- 
 Second System of Units, or F. Jenkin's Electricity and Magnetism. 
 
ELECTROMETERS. 2-11 
 
 only form which can be used for obtaining with the aid of 
 photography a continuous record. At Kew Observatory a 
 quadrant electrometer is now in action ; and it is very much 
 to be desired that at every meteorological observatory atmo- 
 spheric electricity should be made a subject of continuous 
 observation. It seems strange that it has not already re- 
 ceived more attention ; or rather that but few observatories 
 have hitherto given the subject any attention whatever. 
 [Some of the photographic traces obtained at Kew with an 
 older form of reflecting divided ring electrometer, exhibited 
 in the Loan Collection, were shown and described.] 
 
 I have just one other class of electrometers to refer to. 
 Electrometers of this class are called " attracted-disc elec- 
 trometers." The first instruments of this kind were made 
 by Sir W. Snow Harris, but they were in many ways very 
 imperfect, and, as I had occasion to remark yesterday, his 
 interpretation of the results which he obtained led him into 
 the error of disputing the truth of the laws of Coulomb. 
 
 [The diagram (Fig. 8) shows an electrometer of Snow 
 Harris. As exhibited, it is connected with a Leyden jar, j, 
 which is to be tested, while the Leyden jar is being charged 
 from an electric machine through the well-known " unit jar " 
 L.] In this electrometer the attraction between two plates 
 a and d, one of which, a, is insulated and electrified, is 
 determined by weighing, as in a common balance. The 
 plate a, being electrified, attracts d, and the attraction is 
 counterbalanced and weighed by putting weights into the 
 pan P of the balance. Now if we have two circular elec- 
 trified plates attracting each other, the "lines of force" 
 between the plates being straight lines, and if we measure 
 the areas of the plates, the distance between them and the 
 force of attraction, it is easy to deduce the difference of 
 potentials between them in absolute measure. This was 
 practically what Snow Harris attempted. 
 
 As I have said, however, his arrangement was very 
 imperfect. The movable plate 'd is in no way protected 
 from inductive influence, and of the nature or extent of that 
 influence there is no possibility of taking any account. 
 Moreover, the arrangement of the instrument is not such 
 as to enable us to consider the lines of force between the 
 attracting plates to be sufficiently nearly straight lines. 
 Sir William Thomson, in taking advantage, for an absolute 
 
242 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 electrometer, of the principle of weighing the attraction 
 between two discs, has remedied both these defects. The 
 absolute electrometers are not very portable, and thus 
 there is not one to be seen in the Loan Collection. These 
 diagrams will, however, I hope, enable you to understand 
 the principles of the instrument. 
 
 If we have two electrified discs attracting each other, their 
 
 t'iG. S. 
 
 radii being two or three times as great as the distance 
 between them, it can easily be shown that there is a region 
 about the middle where the lines of force are sensibly 
 straight lines. And this condition will also be practically 
 realised by such an arrangement as is show r n in Fig. 9. 
 A and B are two insulated electrified plates. In the middle 
 of B a circular hole is cut, which is filled up by a very light 
 movable disc of aluminium, c, borne at the end of the lever L. 
 
ELECTROMETERS. 243 
 
 The disc c, when " in position," is placed so that its lower 
 face is as nearly as possible in the same plane with the lower 
 surface of the "guard-ring" B. The interstice between B 
 and c is extremely narrow in comparison with the diameter 
 of c and with the distance between the plates. This is 
 the electrical part of the absolute electrometer. 
 
 The lever L is pivotted on a torsion wire, which is stretched 
 between two insulated metallic pillars PP. Q is a counter- 
 poise. At the end of the lever L there is an index hair ; 
 and a lens I is placed so as to view the position of the hair 
 relatively to two black dots on an index plate. When the 
 hair is between the two black dots the plate c is in position, 
 
 with its lower surface in the plane with the lower surface 
 of the guard-ring B. The guard-ring B is kept metallically 
 connected with the metal pillars PP, and thus with the lever L 
 and with the plate c, which, however, is not in direct contact 
 with B, but is perfectly free to move. The lever L is hung 
 so that when A and c are at the same potential the counter- 
 poise preponderates, and the hair at the extremity of L 
 rises above the sighted position ; and before each series of 
 experiments with this electrometer the force required to 
 bring the hair to the sighted position is determined by 
 placing a small weight on c and a rider on the arm L. 
 When this has been done the weights are removed. But 
 when A and c are at different potentials the attraction 
 
 B 2 
 
244 LECTURES TO SCIENCE TEACHERS. 
 
 between them draws the plate c down, and the force with 
 which c is attracted downward depends on the difference 
 of potentials between the plates A and c, and on the distance 
 between them. The plate A can be raised or lowered by 
 means of a screw, and, when an experiment is being made, 
 its distance from c is adjusted till the index hair is in 
 the sighted position. When that is the case it is known 
 that the attraction between the plates is equal to the force 
 of gravity on the weight previously determined. 
 
 Now it was shown by Sir William Thomson in one of his 
 earliest papers on electricity, a paper in which he considered 
 the validity of the objections raised by Snow Harris against 
 the laws of Coulomb, that in a case such as we have been 
 considering, where the lines of electric force are straight 
 lines between two attracting discs, the force, F, of attrac- 
 tion, will be given by the formula 
 
 in which v is the difference of potentials of the two plates, 
 s the surface of the smaller, and D the distance between 
 the two. From this we obtain at once 
 
 Hence if we measure the area s, the distance D, and the 
 force F, which, as I have said, is done by the previous 
 weighing, we can at once calculate v ; and it is to be 
 noticed that if F is expressed in absolute units of force, 
 the numerical value of v deduced by this formula will 
 be obtained in absolute electrostatic units of potential. 
 
 In working practically with the absolute electrometer, how- 
 ever, Sir William Thomson found great difficulty in measur- 
 ing D, the distance between the plates, with sufficient 
 accuracy. He therefore adopted a mode of employing the 
 instrument somewhat different from that which I have in- 
 dicated. The plate A is connected with a separate electro- 
 meter, and with a Leyden jar and replenisher, and is main- 
 tained at a constant high potential. To measure the 
 potential of any conductor, the plate c is first connected 
 with the earth, the plate A is adjusted till the hair is in the 
 sighted position, and a reading taken, called the " earth 
 
ELECTROMETERS. 
 
 245 
 
 reading." Let D be the distance of A from c for this adjust- 
 ment. The conductor to be tested is then connected with 
 c, and by moving the plate A the hair is again brought 
 to the sighted position. Let D' be the new distance. If v' 
 
 FIG. 10. 
 
 and v be the potentials of the earth and of the tested 
 conductor, we see at once that 
 
 V'-V=(D'-D) 
 
 or if v, the potential of the earth be taken as zero, which 
 
246 LECTURES TO SCIENCE TEACHERS. 
 
 is usually done, we have the potential of the conductor 
 tested _ 
 
 V'= (V - D) ' 
 
 S 
 
 Now it is plainly unnecessary to determine the absolute 
 distances D 7 and D. The difference, or the change of dis- 
 tance from a given position of the lower plate A, is all that 
 is required. This Sir William Thomson has found it easy 
 to measure with sufficient exactness by means of a micro- 
 meter screw. 
 
 The form of absolute electrometer that I have now 
 described has been recently superseded by the instrument 
 shown (Fig. 10). The principles of the instrument are 
 precisely the same as those that I have been explaining. 
 The plates A and B, Fig. 10, correspond to those marked A 
 and B, Fig. 9. The plate c of the instrument just described 
 is, in the new instrument, hung somewhat differently, 
 being suspended on three springs somewhat like coach 
 springs. The movable plate and its suspending springs 
 are covered and protected from inductive influence under 
 a cylindrical cover D, of which a part is shown displaced in 
 the diagram. The plate A is moved up or down by a 
 micrometer screw moved by a milled head c' below. The 
 upper part of the glass cover is a Leyden jar, with which 
 is connected a gauge J, similar to that of the quadrant 
 electrometer, and R is a replenisher for maintaining the 
 Leyden jar at a constant potential. In this instrument, 
 however, it is the guard-ring B and the suspended disc that 
 are maintained at a constant potential ; while the conductor 
 to be tested is connected with the electrode E, which is 
 connected with the plate A by the spiral spring R. There 
 are other particulars of which I should like to speak to you, 
 had we the instrument itself before us, and did our time, 
 which I perceive is almost gone, permit. 
 
 Lastly, I must show you a very beautiful instrument, the 
 Portable Electrometer (Fig. 1 1). A few words will suffice for 
 its description. In the bottom of a cylindrical glass jar, 
 of which a considerable part is coated with tin-foil, so as to 
 form a Leyden jar, there is fastened a circular brass plate, 
 with a movable aluminium square, and long index arm, pre- 
 cisely the same as the gauge of the quadrant electrometer 
 
ELECTROMETERS. 
 
 2 4' 
 
 (Fig. 7), only that it is inverted. This plate is kept 
 electrically connected with the interior coating of the jar ; 
 and the jar is kept at a high potential. There is a hori- 
 zontal brass plate A, which is borne on a glass column at its 
 middle, and insulated from everything but the spiral spring 
 r. This spiral spring is connected to an electrode which 
 passes, insulated, out through the top of the case of the 
 instrument, under the " umbrella " D ; and by means of this 
 electrode the plate A may be connected with any conductor 
 whose potential it is desired to measure. The plate A can 
 
 FIG. 11. 
 
 be raised and lowered. For this purpose the glass column 
 by which it is suspended is attached to a hollow nut m t 
 which works up and down on a vertical micrometer screw 
 turned by the milled head c. A vertical scale [not shown 
 in the diagram] and the horizontal graduated circle con- 
 nected with c, serve to read off changes of distance of A 
 relatively to the attracting plate G ; p is a mass of pumice, 
 on which drops of the strongest sulphuric acid are poured, 
 for the purpose of keeping the interior of the case dry. 
 
 The portable electrometer is used in a way very similar 
 to that in which the absolute electrometer is used. The 
 
248 LECTURES TO SCIENCE TEACHERS. 
 
 electrode connected with the plate A is first connected with 
 the outside casing of the instrument, that is practically with 
 the earth, and the micrometer screw is turned till the index 
 hair is in the sighted position. The earth reading is then 
 taken. Next the electrode is connected with the conductor 
 to be tested. If the conductor is at a different potential 
 from the earth, the attraction of the plate A for the plate 
 G is altered. The micrometer screw is again turned till the 
 index takes the sighted position. From the difference of 
 the two readings the potential of the conductor tested 
 may be deduced. This instrument, however, requires a 
 preliminary determination of constants in a way similar 
 to that which I have already indicated for the quadrant 
 electrometer, in order to enable us to reduce the numbers 
 deduced from the scale-readings to absolute measure. 
 
 The portable electrometer is admirably fitted for obser- 
 vation of atmospheric electricity. As I show you it now it 
 is ready for that purpose. To the electrode of the plate A 
 a long stiff wire is connected, which bears at its point a 
 slow-burning match of blotting-paper, soaked in nitrate of 
 lead, and then dried. When I light the match the heated 
 particles rushing off from it quickly bring the potential of 
 the plate A to be the same as that of the atmosphere just at 
 the point where they are rushing off. By turning the 
 micrometer screw I quickly adjust the position of the plate 
 A till the index is in the sighted position. 
 
ON THE APPARATUS RELATING TO 
 VEGETABLE PHYSIOLOGY. 
 
 EY SYDNEY H. VINES, B.A., B.SC. ; FELLOW OF CHRISTS 
 COLLEGE, CAMBRIDGE. 
 
 IN describing and explaining to you the construction 
 and the use of the instruments relating to Vegetable 
 Physiology which have been contributed to the Loan 
 Exhibition, I will follow the order in which they are 
 arrang'ed in the Catalogue. 
 
 The first object to which I would draw your attention 
 is this sketch (No. 3904 in Catalogue) exhibited by Dr. 
 Velten, Physiologist to the Institute for experiments 
 relating to Forestry in Vienna of an apparatus devised 
 for the investigation of the influence of temperature upon 
 living organisms. It consists of a box of zinc with 
 double walls, in the roof and sides of which panes of 
 glass are fixed, by means of which all that goes on within 
 can be observed, and in the sides are openings through which 
 the hands cased in india-rubber gloves, to prevent any 
 sudden change of temperature may be introduced into 
 the interior. The space between the walls is filled with 
 a liquid (water or oil), the temperature of which can be 
 raised by applying heat below, or lowered by placing a 
 refrigerator in the wooden case made to receive it. 
 
 The use of this apparatus will, I think, be most clearly 
 demonstrated by a description of some experiments which 
 might be performed with it. In performing such experi- 
 ments we must be most careful to arrange that the 
 phenomena which we are to observe shall be dependent 
 only upon that one agent the action of which we are 
 investigating, namely, temperature. "We must therefore 
 
250 LECTURES TO SCIENCE TEACHERS. 
 
 prevent as far as possible the action of other agents upon 
 the organism under experiment, and this may be done 
 either by entirely removing the agents in question, or by 
 rendering this action constant throughout the experiment. 
 In the present case the disturbing agents are light and 
 moisture. The action of the former may be prevented by 
 keeping the apparatus in the dark, and of the latter, by 
 maintaining the atmosphere within the case at its point of 
 saturation. 
 
 The first and simplest experiment which we might 
 perform would be one in which a plant should be exposed 
 to a constantly increasing temperature, and we should find 
 that when a certain degree in temperature was reached, 
 the plant would die. In the next place, we should find 
 that, if a plant were exposed to a constantly decreasing 
 temperature, after a time death would ensue. From these 
 experiments we should learn that plant life can be main- 
 tained only within certain limits of temperature and by 
 repetitions of these experiments we should be able to 
 determine thesa limits with accuracy. Roughly speaking, 
 they are O C. (32 F.) and 50 C, (122 F.) (Sachs). 
 
 The next step would be to investigate the relation 
 existing between each one of the phenomena of plant-life 
 and the temperature of the surrounding medium. Suppose, 
 for instance, that we are investigating the effect of tempera- 
 ture on growth. Taking the Pea as the subject of experi- 
 ment, we should find that it would not grow at all at a tem- 
 perature lower than 6 C., that the rapidity of its growth 
 would increase with every rise of temperature up to 26 C. 
 that any further rise of temperature would be attended 
 with diminished rapidity of growth, and that, at a tempera- 
 ture of about 40 C., growth would be entirely arrested 
 (Koppen). 
 
 A series of such experiments upon any plant would 
 show that all its functions are affected by temperature in 
 the manner just described with reference to growth. We 
 should find that, for each function, there were certain 
 definite limits of temprature within which that function 
 could be performed, and that between these limits there 
 was a degree of temperature which corresponded to the 
 maximum activity of the function in question. 
 
 The next apparatus in the catalogue (No. 3935) is but a 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 251 
 
 special application of the preceding to the investigation of 
 the relation existing between the process of germination 
 and temperature, devised by Prof. Cohn, Director of 
 the Institute of Vegetable Physiology in the University 
 of Breslau. It consists of a chamber with double walls 
 made of tin the space between them being nearly filled 
 with water within which are several tin trays on each of 
 which are several small dishes of porous earthenware. The 
 seeds, after having been soaked for twenty-four hours 
 in water, are placed in the earthenware dishes, water being 
 poured into the tin trays in order to maintain the seeds 
 in a sufficiently moist state. Heat is then applied to the 
 chamber by means of a small regulated gas flame, and the 
 temperature is raised to a degree which previous experi- 
 ments have shown to be the most favourable to the process 
 of germination. 
 
 The use of an apparatus of this kind to the physiologist 
 is obvious. By its assistance he is enabled to control all 
 the conditions upon which the process of germination 
 depends, and he can ensure a supply of material at any 
 time for the investigation of this process either from a 
 physiological or from a morphological point of view. But 
 its usefulness is by no means limited to the study of the 
 process of germination as it occurs in the seeds of the higher 
 plants. It is extremely useful for the purpose of culti- 
 vating various kinds of fungi, since it affords all the 
 conditions necessary for their growth, and by this means, 
 much of the life-history of these fungi may be brought 
 under observation. 
 
 Let us consider for a moment what lessons in Vegetable 
 Physiology this apparatus teaches. To the seeds germin- 
 ating, or to the moulds growing within it, it affords heat, 
 air, and moisture, whilst it shuts out the light from them. 
 Hence we may infer that heat, air, and moisture are 
 essential to the germination of seeds, and to the growth of 
 moulds, whereas the presence of light is unnecessary. 
 
 With this apparatus we can further ascertain what 
 degree of temperature is most favourable to the germin- 
 ation of a seed or to the growth of a fungus, and between 
 what limits of temperature these processes will occur. 
 For instance, it has been found that wheat or barley will 
 germinate at a temperature so low as 5 C., and the highest 
 
252 LECTURES TO SCIENCE TEACHERS. 
 
 temperature at which the seeds of these plants have been 
 observed to germinate is 38 C. the temperature at which 
 germination is most active being from 20 to 25 C. 
 Experiments of this kind performed upon the seeds of a 
 variety of plants show that the range of temperature 
 through which germination may take place, extends from 
 to 50 C. 
 
 The next piece of apparatus (No. 3936) is quite classical 
 in the history of Vegetable Physiology. It is constructed 
 and exhibited, like the preceding, by Prof. Dr. F. Cohn, of 
 the University of Breslau, and is a modification of the 
 original apparatus invented by Knight in the year 180G. 
 It consists of a large tin box in which is an axle bearing 
 a water-wheel at one end and a disc of cork at the other. 
 A stream of water is introduced through the roof of the 
 box which sets the water-wheel in motion, causing, at the 
 same time, rotation of the disc of cork, into which are 
 fixed several pins, each of which transfixes the cotyledons 
 of a pea. This apparatus has been in action for the last 
 few days, and as you see, tho peas have germinated. I 
 would ask you particularly to notice that in each case the 
 young stem has grown inwards towards the centre of the 
 axle, whereas the young root has grown outwards in the 
 opposite direction. 
 
 Knight was led to invent this apparatus from a con- 
 sideration of the fact that the young root of a germinating 
 seed always tends to grow towards the centre of the earth, 
 whatever the position of the seed may be, whereas the 
 young stem always grows in the opposite direction. 
 Duhamel had already shown that, if the position of a seed 
 be reversed several times whilst it is germinating, the 
 young root and stem soon resume the original direction of 
 their growth after each change of position. These pheno- 
 mena were ascribed at that time to the action of gravity 
 upon the growing tissues, but these views were purely 
 hypothetical, resting upon no experimental basis : it, 
 remained for Knight to provide the necessary evidence. 
 
 By means of this apparatus, the stem and root of the 
 young plant are exposed to precisely similar conditions of 
 light, moisture, air, and temperature, the action of gravity 
 is eliminated, and the action of a considerable centrifugal 
 force is introduced. Under these circumstances we must 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 253 
 
 admit that it is under the influence of this force that the 
 root grows centrifugally, and the stem centripetally that 
 is to say, that the direction of growth of each of these 
 organs depends upon a relation existing between a purely 
 physical force and their growing cells. 
 
 Applying these results to the ordinary downward growth 
 of roots and upward growth of stems, we are at once led 
 to the conclusion that these organs assume their respective 
 directions of growth in consequence of the action of 
 gravity upon their growing cells. 
 
 The mode in which gravity acts upon growing cells 
 may be to some extent rendered intelligible by means of a 
 diagram, (Fig. 1). Suppose A B to be a young plant of 
 
 Fio. 1. 
 
 which A represents the stem and B the root, each consisting 
 of a single cell. At first the position of both is horizontal, 
 but as growth proceeds A turns upwards and B down- 
 wards. This change in direction is due, in the case of A, 
 to the fact that the lower portion of the cell-wail has grown 
 more rapidly than the upper portion, and in the case of 
 B, to the fact that the upper portion of the cell-wall has 
 grown more rapidly than the lower. 
 
 Why the growing cells of young roots should be so 
 affected by physical forces, such as gravity and centrifugal 
 force, that the direction of their growth should follow that 
 of the action of these forces, and why the growing cells 
 of young stems should be affected in a precisely opposite 
 manner, are questions to which no satisfactory reply can 
 be given at present. 
 
 To this behaviour of the growing cells of plants to the 
 action of gravity the name geotropism has been given 
 those parts which obey it in the direction of their growth 
 being termed positively, those which oppose it negatively, 
 
254 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 geotropic. Geotropisin is by no means confined to roots 
 and stems, nor is its negative or positive character in- 
 separably connected with the morphological nature of the 
 organs of plants. It is true that all primary roots are 
 positively geotropic, as well as most secondary roots, but 
 so also are many leaf -bearing axes, and even leaves, as in 
 the case of the cotyledonary sheaths of Allium and other 
 Monocotyledons ; and to these we may add the lamellae 
 forming the hymenium of mushrooms so that we have 
 an assemblage of structures of the most varied morpho- 
 
 FJG. 2. 
 
 logical value, all of which exhibit positive geotropism. On 
 the other hand, all erect leafy axes, petioles, stipes of mush- 
 rooms, conidiophores of moulds, are negatively geotropic. 
 
 We now come to an apparatus (No. 3938) which is very 
 much more complicated than any of those with which we 
 have already become acquainted. It is an apparatus 
 exhibited and constructed by Herr E. Stohrer, of Leipsic, 
 for registering the growth of plants (Fig. 2). A fine 
 thread is attached by one end to the growing internode of 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 255 
 
 the plant under observation, it passes over a pulley, and at 
 its other end a small weight is suspended. It is clear that 
 any elongation of the plant will cause a rotation of the 
 pulley in consequence of the friction of the thread upon it. 
 In close connection with this pulley is a toothed wheel, 
 and any movement of the pulley will cause a rotation of 
 the toothed wheel G. The teeth of the wheel as they rotate 
 come into contact with a small steel spring, and force it 
 against one or other of two small projecting binding screws 
 (a) and (6), from which it springs back when the pressure of 
 the tooth is removed by the further rotation of the wheel. 
 By means of a third binding screw (c) the spring is con- 
 nected with the wire coming from one pole of a galvanic 
 battery C, and wires connect the binding screws (a) and (6) 
 with the two electro-magnets A and B, which are in con- 
 nection with the other pole of the battery. It is evident 
 that when the spring touches the binding screw (a) or (6), 
 a current will be sent through the corresponding electro- 
 magnet A or B, in consequence of which it will attract 
 towards itself a steel rod P, which is supported between the 
 electro-magnets, and will keep it in that position until the 
 current passing through the electro-magnet is broken by 
 the escape of the lever from the pressure of the toothed 
 wheel, when the steel rod P springs back to its place. This 
 steel rod bears a pencil at one end which marks upon a disc 
 D, made to revolve once in twenty-four hours by means of 
 clockwork E. It is clear that every movement of the steel 
 rod bearing the pencil will be registered upon the surface 
 of the disc, and as these movements depend upon the 
 rotation of the toothed wheel, and this finally upon the 
 growth of the plant, the marks of the pencil of the disc 
 will give an indication of the growth of the plant. 
 
 The accompanying sketch (Fig. 3) of a tracing taken 
 upon the disc will make this explanation somewhat more 
 intelligible. Only one half of the disc is represented, and 
 it is divided into twelve segments corresponding to the 
 twelve hours between 6 P.M. and 6 A.M. You will observe 
 that at 6 P.M. the pencil was describing a circle upon the 
 disc, and it continued to do so until just before seven, when 
 the tracing shows a sudden break. This indicates that, 
 owing to the pressure of the toothed wheel, the steel spring 
 has come into contact with the binding screw I, a current 
 
256 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 has been made to pass through the electro-magnet B, and 
 it has attracted the steel rod bearing the pencil. You will 
 also observe that the pencil remains in this new position 
 
 .Fie. 3. 
 
 for some time in fact, for just so long a time as the 
 pressure of the toothed wheel keeps the spring in contact 
 with the binding screw b, for when this pressure is removed 
 the pencil resumes its original position. We see then that 
 each of these depressions which we observe in the tracing 
 corresponds to the time which is required for the rotation 
 of a single tooth of the wheel. 
 
 This is made more evident when the tracing on the disc 
 is tabulated in the manner illustrated by Fig. 4. In this 
 the Arabic figures indicate hours, the Roman figures the 
 depressions in the disc-tracing, or, in other words, divisions 
 of the toothed wheel. 
 
 Fig. 5 is the tracing, similarly tabulated, taken during the 
 twelve hours between 6 A.M. and 6 P.M. 
 
 Fig. 6 is a tabulated arrangement of the two preceding 
 figures, by means of which the curve of the velocity of 
 growth during the twenty-four hours is obtained. In this 
 the smallest velocity is taken as the unit of measurement 
 for the ordinates. 
 
 So much then for the mechanism of this apparatus, and 
 for the mode of using it. Let us now go on to consider 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 257 
 
 what we are to learn from the tracings which we have 
 obtained from it. From an inspection of them we find 
 that growth took place between six o'clock in the evening 
 
 PM. Mv&ruglvt AJH 
 
 6 7 8 9 W 11 IZ 1 2 3 4 5 6 
 
 W 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 11 
 
 m 
 
 JL 
 
 i 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 / 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 > 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 FIG. 4. 
 
 and ten o'clock tha following morning, after which a period 
 of inactivity followed, until, towards four o'clock in the 
 
 1 
 
 \'l 
 
 V 
 IV 
 
 m 
 ii 
 j 
 
 4Jf Xcon, PJk 
 S 7 b S W n IZ 1 Z 3 4 5 6 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 f 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 / 
 
 
 
 
 
 
 
 
 
 
 
 FIG. 5. 
 
 afternoon, the process of growth recommenced; and we 
 find further that the period of growth is divided into two 
 well-marked epochs, the one extending from 7 P.M. until 
 
 8 
 
258 
 
 LECTURES TO SCIENCE TEACHERS. 
 
 3 A.M., the other from 7 A.M. to 10 A.M. We have now to 
 account for these intermissions in the process of growth, 
 and to discover whether they bear any relation to variations 
 occurring in the external circumstances to which the plant 
 was exposed. These external circumstances were (1) a 
 supply of air, (2) a supply of moisture, (3) the action of 
 temperature, (4) the action of light. 
 
 We may fairly assume that of these the first two were 
 invariable during the course of the experiment, and we 
 need not therefore regard them any longer as elements of 
 the problem. It remains for us to consider the action of 
 temperature and of light upon the growth of ^.e plant. 
 
 In discussing the question as to how much changes of 
 temperature have contributed to the production of these 
 tracings, we must recall the general law at which we have 
 already arrived, which indicates the relation existing 
 between temperature and plant life. From this we are 
 able to assert that a slight rise of the atmospheric tempe- 
 rature will stimulate the process of growth, whereas a 
 slight fall will depress it. Now we shall be probably 
 correct in saying that the temperature of the air will 
 gradually sink from 6 P.M. until sunrise (3 A.M.), after 
 which it will gradually rise, until it begins to sink again 
 in the afternoon (4 P.M.). We should therefore expect to 
 lind that the process of growth would be arrested, or at 
 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 259 
 
 any rate would proceed but slowly, from 6 P.M. to 3 A.M., 
 and that then it would go on with increasing rapidity until 
 the afternoon. We find, however, as a matter of fact, 
 that growth has gone on vigorously from 6 P.M. to 3 A.M., 
 and that the tracings support our supposition only in 
 regard to the growth which took place between 7 A.M. and 
 10 A.M. We are forced to the conclusion that the growth 
 of the plant has been influenced by some force having an 
 effect upon it contrary to that of temperature and we 
 must seek this force in the action of light. 
 
 Let us employ a process of reasoning in the investiga- 
 tion of this question, similar to that which we employed 
 in the case of temperature. We may begin by assuming 
 that which already appears to be probable, namely, that 
 the effect of light is to retard growth, and we may go on 
 to see whether or not this hypothesis satisfactorily explains 
 the phenomena. We should expect to find that with dim 
 light (from 6 P.M. to 3 A.M.) the process of growth would 
 commence, or, if already in progress, would go on with 
 increased rapidity ; and we should further expect to find 
 that, after sunrise, the rapidity of growth would dimmish 
 until it was completely arrested. The tracings bear out 
 the first of our expectations with precision, but the epoch 
 of growth extending from 7 A.M. to 10 A.M. is quite 
 contrary to the second. It is evident from these facts 
 that the intermissions of growth are not due to the action of 
 light alone, nor to the action of temperature alone : it 
 remains for us to see whether they are not due to the 
 combined action of these influences. From 7 P.M. to 3 A.M. 
 we have a dim light and a decreasing temperature. Of 
 these conditions, the former is favourable, the latter un- 
 favourable, to the process of growth. If we reflect that the 
 diminution of temperature must be after all comparatively 
 small, we may fairly conclude that the absence of light has 
 had a greater effect than the diminution of temperature, and 
 that therefore the process of growth has on the whole been 
 favoured. From 3 A.M. to 7 A.M. we have a rapidly increasing 
 intensity of light and a slowly rising temperature, and we 
 find, as we might have expected, that the tracing indicates 
 a period of inactivity. From 7 A.M, to 10 A.M. we have 
 again an increasing intensity of light and a rising temper- 
 ature, the latter taking place more rapidly than the former, 
 
260 LECTURES TO SCIENCE TEACHERS. 
 
 where the tracing indicates a short epoch of growth. From 
 10 A.M. to 4 P.M. the influence of a high temperature seems 
 to be entirely overcome by the action of an intense light, 
 so that it is not until the afternoon, when the light begins to 
 wane, that the growth is resumed. 
 
 Hitherto we have regarded only Figs. 4 and 5, which 
 indicate the actual absolute growth of the plant. Let us 
 briefly consider Fig. 6, which shows the relative velocity of 
 growth at different times. The growth was slowest at 
 4 P.M. at 8 P.M. the figure indicates that its rapidity was 
 twice as great then we observe a slight diminution of 
 velocity towards midnight, but an increase soon followed 
 and continued until 10 A.M., when the greatest velocity 
 three times that at 4 P.M. was attained. 
 
 From the discussion of an experiment performed with 
 this apparatus we have derived a confirmation of the 
 general law which expresses the relations between plant- 
 life and temperature, and we have learned that light has a 
 retarding effect upon the process of growth. It may be 
 objected, however, that the intermissions of growth recorded 
 in the tracings may not, after all, be entirely due to the 
 influence of external conditions upon the plant, but that 
 they may perhaps be referred to some inherent property of 
 the growing tissues by virtue of which a certain periodicity 
 in the process is produced. Experiments have been in- 
 stituted in which the external conditions were prevented, 
 as far as possible, from undergoing any variation during 
 their performance, and in these no such intermissions of 
 growth were observed as are indicated in Figs. 4 and 5. 
 
 We may now go on to inquire more deeply into the 
 action of light upon growth, and the first problem which 
 presents itself is to find what rays of the solar spectrum 
 are particularly concerned in it. We might obtain the 
 solution of this problem by exposing the growing parts of 
 a plant successively to the action of the rays of different 
 refrangibility which make up the solar spectrum, and by 
 comparing the rates of growth observed. We should find 
 that the rapidity of growth would gradually diminish as 
 the plant was removed from the red towards the blue end 
 of the spectrum, until it entirely ceased. This mode of 
 experiment is, however, scarcely suitable for the general 
 purposes of demonstration, as it must be carried out with 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 261 
 
 great care, and involves the use of a good deal of appa- 
 ratus. The same results may be more easily arrived at by 
 means of these two flasks which I have on the table, and 
 which were devised by Prof. Sachs, of the University of 
 Wurzburg (No. 3944). They are simply large glass bottles 
 with double walls, the space between the walls of the one 
 being filled with an ammoniacal solution of copper oxide, 
 and of the other, with a solution of potassium bichromate 
 the former being of a blue, the latter of an orange colour. 
 In this way we are enabled to split up the solar spectrum 
 into two halves, the one half of its rays (including part 
 of the green, the blue, indigo, violet, and the actinic rays) 
 penetrates into the cavity of the blue flask the other 
 half (including part of the green, the yellow, orange, 
 red, and dark rays) reaches the interior of the yellow flask. 
 That the properties of these two sets of rays are very 
 different is indicated by the fact that a piece of sensitive 
 paper, such as is used in photography, rapidly becomes 
 darkened when exposed in the blue flask, whereas it is 
 scarcely affected by exposure in the yellow one ; and we 
 may therefore conclude that, in the former case, we have 
 to deal with rays of considerable actinic power, whereas, in 
 the latter, the rays are comparatively inert. Let us see 
 how these two sets of rays affect the process of growth. 
 Here are some young pea-plants which have been grown for 
 some days inside these flasks, and have been supplied 
 equally with air and moisture. It will strike you at once 
 the growth has taken place much more rapidly under the 
 influence of yellow than under that of blue light, for there 
 is a difference of several inches between the heights of the 
 plants. From this we may conclude that the rays of high 
 refrangibility are more active in arresting growth than 
 those of lower refrangibility. 
 
 A careful examination of these two plants will show 
 that they differ not only in the amount, but also in the 
 direction of their growth. You see that this plant grown 
 in the blue flask has become considerably curved in a 
 definite direction, and this was towards the source of its 
 illumination, whereas the plant grown in the yellow flask 
 is nearly straight. This tendency to bend towards the 
 source of light is due, then, to the action of the highly 
 refrangible rays upon the growing cells. These rays have 
 
262 LECTURES TO SCIENCE TEACHERS. 
 
 a more powerful action in retarding the growth of those 
 cells of the stem which are more immediately exposed to 
 them than they have in retarding that of the cells more 
 removed from their influence, and consequently the stem 
 becomes bent towards the source of light. 
 
 To the phenomena resulting from the action of light 
 upon the process of growth the general term heliotropism 
 is applied. This young pea-plant, then, affords an example 
 of heliotropism, and since the direction of its curvature 
 is towards the source of light, the heliotropism is said 
 to be positive. It does happen, however, that in some 
 plants and parts of plants the direction of curvature 
 is away from the source of light (for instance, in the 
 tendrils of Vitis and Ampelopsis and many aerial roots), 
 and under these circumstances the heliotropism is said to 
 be negative. The explanation of this negative heliotropism 
 is by no means satisfactory as yet. Positive heliotropism 
 we said, depends upon a retardation of the growth of the 
 cells of the more brightly illuminated side of the stem, 
 and we might be inclined to go on to say that negative 
 heliotropism depends upon an increased rapidity of the 
 growth of the cells upon the more brightly illuminated 
 side to admit, in fact, that the cells of some plants have 
 their growth diminished by the action of light, whereas 
 the cells of others havs their growth increased. An 
 attempt has been made to explain the negative heliotropism 
 of tolerable transparent structures (e.g. the aerial roots of 
 Aroids) by showing that in such cases the side most 
 removed from the source of light receives a more intense 
 illumination than the side nearest to it in consequence of 
 the refractions which the rays of light undergo after their 
 penetration into the tissues of the organ (Wolkoff), and 
 under these circumstances, the apparent negative helio- 
 tropism of these organs would be merely a special case of 
 positive heliotropism. It is evident, however, that this 
 theory is quite inapplicable to the cases of negative 
 heliotropism occurring in organs which are not transparent. 
 The true explanation of negative heliotropism has yet, I 
 believe, to be discovered. For the present we may perhaps 
 be contented to assume provisionally the hypothesis that 
 growing cells are of two kinds, the cells of the one kind 
 having the rapidity of their growth increased, those of the 
 
APPARATUS OF VEGETABLE PHYSIOLOGY. 263 
 
 other having the rapidity of their growth diminished by 
 the action of light, that there are, in fact, cells which are 
 positively or negatively heliotropic, just as there are cells 
 which are positively or negatively geotropic. 
 
 Growth is, however, only one of the processes of plant 
 life which are influenced by the mechanical action of the 
 more refrangible rays of the spectrum. For instance, the 
 movements of zoospores, 1 as well as those of the motile 
 parts of the higher plants, are all influenced by these rays ; 
 and we may say generally that the more purely physical 
 phenomena of plant life are especially affected by them. 
 
 The less refrangible rays such as would penetrate into 
 the interior of this orange-coloured flask are concerned 
 with the chemical processes of plants. Of the truth of 
 this statement I am unable on this occasion to give you 
 any demonstration. I can, however, refer you to experi- 
 ments which show conclusively that some of the most 
 important chemical processes can only take place in the 
 presence of such rays. Assimilation, for instance that 
 process in which the carbonic acid gas present in the air 
 is decomposed by the green colouring matter of the leaves, 
 and the carbon thus obtained is combined with oxygen and 
 hydrogen derived from water to form starch, whilst at the 
 same time oxygen is liberated can only take place under 
 the influence of these rays of low refrangibility, the yellow 
 rays being particularly active in promoting this process 
 (Draper ; Pfeffer). 
 
 With the apparatus of Prof. Sachs our list closes. 
 The instruments exhibited are few in number and are 
 comparatively simple in construction, but they are of 
 considerable interest, for the application of them has 
 materially contributed to the solution of some of the more 
 important and difiicult problems of Vegetable Physiology. 
 
 1 Since the above was written Prof. Sachs has published a paper,' 
 in which he shows that the apparent heliotropism of zoospores is pro- 
 duced by currents in the water, in consequence of slight differences 
 of temperature in different parts. 
 
ELECTRICAL MEASUREMENTS. 
 
 TWO LECTURES. 
 
 BY PROF. CAREY FOSTER. 
 
 LECTURE I. 
 
 UNDER the circumstances of these Lectures I think it will 
 probably "be most useful, if, instead of attempting to 
 describe the details of the construction or use of special 
 instruments, I try to explain as well as I can in the time 
 available, the principles upon which the use of instruments 
 of any general class depends. You are aware that the 
 subject of the present lecture, and the one to follow next 
 week is Electrical Measurement, and I shall to-day speak of 
 measurements which have relation to Statical Electricity or 
 electrostatic measurements. Here the magnitudes which 
 have to be measured are the Quantity of electricity, the 
 Potential of the electricity, and the electrical Capacity of the 
 bodies which may be charged. Then we have also the Energy 
 represented by a quantity of accumulated electricity or a given 
 electric charge. These are the fundamental magnitudes with 
 which we have to deal in electrostatics ; but there are various 
 secondary matters which may be made a subject of measure- 
 ment, for instance electrical Density, which is the quantity of 
 electricity per unit of volume. If we have any electrified 
 substance the quantity of electricity in unit of volume of that 
 body is the density of the charge ; if we have a charged con- 
 ductor, then, as is well known, the electric charge is confined 
 
ELECTRICAL MEASUREMENTS. 265 
 
 to the surface so that we have not to deal with volume in the 
 case of conducting bodies but with surface only. The super- 
 ficial density is the quantity of electricity per unit of surface. 
 Thus electrical Density is a subsidiary magnitude connected 
 with Quantity. Then in connection with Capacity there is a 
 magnitude which may be regarded as a subsidiary one. You 
 are well aware that the capacity of a given conductor, of 
 this cylinder for instance, that is the quantity of electricity 
 which it must take up in order to have a charge of given 
 potential, depends partly on the form and partly on the 
 dimensions, but also on the distance of surrounding bodies 
 and on the nature of the substance which separates the 
 conductor we are discussing from the surrounding . bodies. 
 There is then a property of the surrounding medium which 
 has something to do with determining the capacity of a given 
 conductor. This cylinder for instance is surrounded by 
 conductors such as the walls of the room, the ceiling, the 
 table, and so on, as well as by other nearer bodies, and it is 
 separated from these conductors by air. If we were to 
 replace the air by another insulating medium, the electrical 
 capacity of the cylinder would be changed ; and that 
 property of the surrounding medium which enables it to 
 produce an effect on the capacity of a given conductor is 
 called the specific inductive capacity of the medium or the 
 dielectric co-efficient, these being synonymous names for the 
 same property. 
 
 The measurement of these different magnitudes constitutes 
 the problem that we have to discuss in talking of electro- 
 static measurements. In the case of any measurement we 
 may adopt what we may call a direct process, or an indirect 
 one, and to make more clear what I mean by this distinction 
 I may remind you of familiar processes in other cases. I 
 call a direct measurement one such as an ordinary measure- 
 ment of length. If we want to measure the length of this 
 table we take a foot-rule or any other standard we may agree 
 upon, and by measuring off this standard length over and over 
 again we estimate the length of the table in terms of the 
 adopted unit : but in the case of electricity such a process as 
 this is not available. One cannot take a standard quantity 
 of electricity and compare this by a direct process with any 
 other quantity that we want to measure. We have to adopt 
 a process less direct, in which we observe the effect which the 
 
266 LECTURES TO SCIENCE TEACHERS. 
 
 quantity of electricity which we have to measure can produce, 
 and compare this with the magnitude of the similar effect 
 produced by the standard quantity. This rather abstract 
 way of stating the case will be more intelligible if I give you 
 an example. In the case, for instance, of heat we are pretty 
 much in the same circumstances as in relation to the measure- 
 ment of electricity. We cannot take a unit of heat like a 
 unit of length and compare it with other quantities. We 
 have to produce some effect by the quantity of heat we want 
 to measure. We may for instance employ it to raise the 
 temperature of water from the melting-point of ice to one 
 degree and estimate the quantity of heat by the amount of 
 water whose temperature can so be raised from zero to 1 C, or 
 we may use the heat in melting ice and estimate the quantity 
 of heat by the quantity of ice that it can melt, the unit of heat 
 being of course either that quantity which can raise a unit 
 mass of water one degree, or that which would melt a unit 
 mass of ice. 
 
 So in the case of electricity, we may adopt as a standard 
 quantity, as the unit of electricity, the quantity which can 
 produce some definite effect, and then to measure other 
 quantities we find out how much of the same effect they can 
 produce. If we want to get what we may term an absolute 
 measure some such process as this is always necessary, but 
 where what we require is simply to compare one quantity 
 with another without requiring to know absolutely what 
 either of them is if we want to know for instance that one 
 quantity is 100 times another without knowing how much 
 this one is for mere comparisons of that sort, many other 
 processes are available : for instance, we may adopt, as 
 giving a comparative measure of two quantities of electricity, 
 the number of times which some definite operation has to be 
 repeated in order to produce the one quantity or the other. 
 The number of turns of the handle of a given machine for 
 instance, might be taken as a measure of the quantity of 
 electricity which is employed in a particular experiment. If 
 we turn the handle 10 times we produce a definite quantity : 
 if we turn it 20 times we produce twice as much, when the 
 machine is in the same condition ; but as all of you know who 
 have worked with electrical apparatus, one day a machine 
 will give us much larger quantities than it will another day, 
 when employed in the same manner, so that such a method 
 
ELECTRICAL MEASUREMENTS, 267 
 
 would not be available for comparisons except in experiments 
 made very quickly one after the other. Again, the Electro- 
 phorus:, every time we lift the cover and take a spark from 
 it, gives us the same quantity of electricity ; the cover always 
 comes up with the same quantity, so that we might charge a 
 conductor from the cover of the electrophorus by giving it 
 several sparks one after the other, and we might estimate 
 the charge by countiug the number of sparks which were 
 given. That at first sight seems, as far as it goes, a perfectly 
 satisfactory method of measurement ; but, on consideration, 
 you will see it is not quite so good as it appears, because 
 although the electrophorus possesses each time the same 
 quantity of electricity, it does not give the same quantity 
 each time to the conductor with which we put it into contact. 
 To take a definite example, suppose that when I let the 
 electrophorus touch this sphere it gives up half its electricity 
 to the sphere : then I charge it again and I put it in contact 
 again, it does not give up so much as before. To begin with> 
 it came in contact with an uncharged sphere, but now it 
 comes in contact with a sphere which has already half as much 
 electricity as the electrophorus itself, so that at each repeti- 
 tion of the process we give a smaller and smaller quantity to 
 the sphere, and you will very easily see that on repeated 
 contacts the quantities of electricity imparted to a good con- 
 ductor diminish according to the terms of a decreasing 
 geometrical series. It is clear that the more nearly the 
 potential of the sphere agrees with the potential of the electro- 
 phorus, the less is the quantity of electricity which passes 
 from one to the other ; and the same thing applies in the case 
 of an electrical machine. If we connect a machine with a 
 Leyden jar or with a number of Leyden jars, one turn of the 
 handle will cause a certain quantity of electricity to pass into 
 the jar ; another turn will cause almost the same quantity 
 to pass in, but you know that after a time the jar gets as 
 strong a charge as the machine we employ can give to it, and 
 then we may go on turning the machine as much as we like, 
 and the charge in the jar does not increase. Hence the 
 proportionality between the charge in the Leyden battery and 
 the number of turns of the machine only holds good at first. 
 If we go on turning we come to a time when we do not increase 
 the charge at all. The quantity produced is however exactly 
 proportional to the number of turns, if the machine remains 
 
268 LECTURES TO SCIENCE TEACHERS. 
 
 in the same state. If instead of charging a battery or other 
 insulated conductor we allow the charge to pass off to the 
 earth, we get for each turn of the handle a definite number 
 of sparks, each spark corresponding to the passage of a 
 definite quantity of electricity. The quantity produced by 
 each turn of the machine is definite, but the quantity which 
 will go into the battery becomes smaller and smaller as the 
 quantity already in the battery increases. Another way of 
 measuring out the charge for a Ley den battery or any such 
 apparatus is by the employment of a unit jar, one familiar 
 form of which I have here upon the table. The quantity 
 which the second jar receives is measured by the number of 
 sparks which pass into the knob connected with the first, 
 but the point I want to draw attention to is, that the 
 sparks indicate that a definite quantity has passed into the 
 second jar. By the time the first spark comes, a certain 
 quantity has passed into this jar and by the time the second 
 spark comes the same quantity has passed in again ; so that 
 the action of the unit jar is comparable to the case of 
 lading water into a cask out of a measure of definite capacity. 
 The spark is the signal indicating that a certain definite 
 quantity has gone in. It is not that the electricity goes in 
 at the moment of a spark, but the number of sparks counts 
 the number of times which the first jar is emptied into the 
 second, so that when we want to give three units to 
 the second jar, we ought to break the connection as soon as 
 the third spark has passed, not to go on until the fourth 
 spark is nearly occurring. 
 
 All such measurements as these I have referred to are 
 merely comparative ; if we want to know, not merely what 
 is the proportion between one quantity of electricity and 
 another, but what the actual quantity is in any particular 
 case, then we must employ a method founded upon the 
 action which the given quantity can produce, and com- 
 pare this with the amount of the same kind of action which 
 the adopted unit of electricity could produce. The most 
 obvious way of getting an absolute measure of quantities of 
 electricity is founded upon the law established by Coulomb, 
 which applies to the force exerted between two quantities of 
 electricity. If we have two equal quantities of electricity, 
 represented each by q then they attract each other if they 
 are of an opposite kind or repel each other if they are 
 
ELECTRICAL MEASUREMENTS. 269 
 
 similar, with a force represented by this formula / = ^. 
 
 The force is numerically equal to the product of the two 
 charges, or to the square of either of them if they are equal, 
 divided by the square of the distance between the points at 
 which they are collected, q stands for the amount of the 
 charge or quantity of electricity, r for the distance between 
 the charges and /for the force. Heading the relation differently 
 we have q = r\/f; or the measure of the charge is the 
 product of the distance into the square root of the force 
 exerted between the two charges. This formula implies 
 that the quantity of electricity which, placed at unit distance 
 from another equal quantity, repels it with unit force, is 
 taken as the standard quantity of electricity ; so that the 
 unit charge or absolute unit of electricity is the charge 
 which repels an equal charge at unit distance (say one 
 centimetre) with unit force (one dyne). The dyne is the 
 most convenient unit of foree ; it is that force which can in 
 one second give to a mass of one gramme, a velocity of one 
 centimetre per second. But that is not a matter which is 
 essential to our present discussion. You may take what unit 
 you please as the unit of force, but the unit of electricity 
 will depend on that. 
 
 To return to what I was saying : in order to apply practically 
 a system of measurement founded upon this principle, it is most 
 convenient, although not absolutely essential, that the charges 
 that we want to measure should be situated upon spherical 
 conductors, little spheres of metal, or spheres with metal 
 surfaces. The advantage of this form of a charged body is 
 this, that the electricity distributes itself uniformly on the 
 surface of a sphere if that is- at a sufficient distance from 
 other bodies, and this uniformly spherical layer of elec- 
 tricity acts upon other electricity as though it were con- 
 centrated at the centre of the sphere ; so that we may in this 
 case regard the quantities which exist upon the spheres as 
 concentrated at their centres. If they are at a very great 
 distance apart that is strictly true ; if at a moderate distance 
 apart it is nearly true. The apparatus in most general use 
 for this purpose is the Torsion Balance which is fully 
 described in the ordinary Text-books on Electricity, and of 
 which I have an example here. In using it, a charge of 
 
270 LECTURES TO SCIENCE TEACHERS. 
 
 electricity is given to the fixed ball ; then if the movable 
 ball is of the same size as the fixed one, the charge of the 
 fixed ball is divided between the two in equal proportions. 
 The balls then repel each other with a force proportional to 
 the square of the charge upon either of them, that is, pro- 
 portional to one fourth of the square of the original charge. 
 If we measure the force exerted between the two balls each 
 of them charged with half the original quantity put into the 
 apparatus, which we can do by observing the extent to 
 which the suspending fibre is twisted and determining its 
 coefficient of torsion, and measure the distance between 
 the balls, which we can do by observing the angle between 
 them and the length of the movable arm, we get the two 
 factors which determine the quantity upon each ball. Then 
 the square of that is equal to one-fourth the square of 
 the original charge. The details of the use of the apparatus 
 would call for a good many remarks if we had time to enter 
 upon it, but that I cannot do now. 
 
 So far as I have spoken of this apparatus it enables us to 
 measure the quantity of electricity which we put into it 
 upon the fixed ball, but that does not yet give us a method 
 of measuring quantities of electricity in general. If I wanted 
 to measure, for instance, the charge of this sphere, I still 
 might do it by means of this apparatus. I should take the 
 fixed ball out of the balance, let it touch the sphere, and then 
 put it in and measure the charge which it has got. The 
 charge taken from the sphere would bear to the charge that 
 the sphere had to begin with, a definite relation which is not, 
 as one might suppose at first sight, the ratio of the surfaces of 
 the spheres. That you will easily see in this way : If we touch 
 this large sphere with a small one, you might suppose that 
 the charge of the large sphere would be divided between the 
 two in the ratio of the surfaces of the large sphere and of the 
 -small one ; but a little consideration shows that that is not 
 the case, for the density of the charge upon the small sphere is 
 greater than the density upon the large sphere. It varies from 
 one part to another, but the average density on the small one 
 is considerably greater than the average density upon the large 
 one. so that the small sphere takes under these circumstances a 
 larger proportion than corresponds merely to its surface, and 
 the ratio in which the electricity divides between the two 
 spheres bears a complicated relation to their sizes. But tables 
 
ELECTRICAL MEASUREMENTS. 271 
 
 have been calculated by reference to which the proportion in 
 which the electricity divides can generally be easily ascertained. 
 Such facts, however, are applicable only to the case of the 
 partition of electricity between spherical conductors ; but we 
 may make any conductor virtually spherical by putting it 
 inside a hollow sphere. If I charge this ball and then hold 
 it inside this jar by an insulating fibre, the outside of the 
 jar, although the ball does not touch it at all, and there is 
 perfect insulation between them, assumes a charge which is 
 just equal to the charge of the ball ; the inner surface of the 
 jar assumes a charge which is equal but opposite to the charge 
 of the ball, and the outer surface gets a charge which is equal 
 and similar to that of the ball ; so that even without any actual 
 contact we get a charge outside the jar equal to the charge of 
 any body which is inside. It does not matter whether contact 
 takes place or not. A charged body inside a perfectly closed 
 conductor causes on the outside of that conductor a charge 
 exactly equal to its own. If we had better insulation it 
 would be easy by a simple experiment to prove this, but I am 
 afraid I could not in the present state of the atmosphere, and 
 with the probable dampness of the supports, make the 
 experiment in a satisfactory way, and therefore I will not 
 attempt it. The experiment is this : You charge the ball 
 and put it inside the jar without letting it come in contact. 
 Then you touch the outside of the jar, and so render it 
 neutral. If you now take out the ball without allowing 
 contact to occur, although the whole apparatus is neutral as 
 long as the ball is inside, it is not neutral when the ball is 
 taken out ; but if before removing the ball you let it touch 
 the jar which should be done without interfering with 
 the insulation of the jar and then take it out, you will 
 find the apparatus is still neutral. This shews that the 
 electricity which escapes to the ground, when the jar is 
 touched while the ball hangs inside, is equal in quantity 
 to, and of the same kind as, the charge of the ball ; for, 
 after its escape, the jar is rendered permanently neutral by 
 receiving the electricity of the ball. On the other hand, if 
 the jar remains insulated while the ball is inside it, its outer 
 surface retains the electricity which, when it is uninsulated, 
 passes away to the earth, that is, it retains a charge equal 
 and similar to that of the ball. However, independently of 
 any particular experiment to prove the point, this is a prin- 
 
272 LECTURES TO SCIENCE TEACHERS. 
 
 ciple of perfectly general application, that if we have one or 
 any number of electrified "bodies completely inclosed in a 
 hollow conductor which is insulated and possesses no electri- 
 fication independent of the bodies inclosed, the outside of the 
 conductor possesses a charge equal to the sum of all the charges 
 which are inside. Therefore if this conductor is a sphere we 
 have on the outside of that sphere a quantity of electricity 
 exactly equal to all that is inside, and we may measure that 
 quantity by allowing it to be divided between the hollow 
 sphere and a small external one which can be put into the 
 balance. Thus in a very great number of cases the measure- 
 ment of the quantity of electricity in any electrified body can 
 be reduced to the measurement of the electricity of a charged 
 sphere. 
 
 "We have next to speak of Density, or the quantity of 
 electricity on unit of surface. As you are aware, the general 
 method of measuring this is to touch the conductor, on which 
 we want to know the density by a small plane conductor. 
 If the conductor with which we touch the body to be tested 
 is small enough, the outer surface of the plane conductor 
 which may conveniently be a small disc of gilt cardboard 
 becomes, for the time being, practically part of the surface of 
 the body to be examined, so that the quantity of electricity 
 which was previously upon the part of the surface covered 
 by the gilt cardboard is transferred to the cardboard. When 
 we take this away we carry off the quantity of electricity 
 previously on that part of the surface, and can measure it 
 in a torsion-balance ; we can also ascertain the amount of 
 surface upon which it was; then, having a measured 
 quantity and measured surface, the ratio of quantity to surface 
 gives the average density. There is another method which 
 may be employed in special cases for determining the 
 density of a charge. It is founded on this principle, which I 
 have not time to prove, but which is probably known to many 
 of you, that the electrical force just outside a charged con- 
 ductor is equal to four times TT (in its ordinary meaning as 
 representing the ratio of the circumference of a circle to the 
 diameter) multiplied by the electrical density. 
 
 In speaking of the electric force just outside a conductor 
 I mean this : Imagine a unit of electricity close to the surface 
 of the conductor, and suppose it can be put there without 
 disturbing the actual distribution of electricity previously 
 
ELECTRICAL MEASUREMENTS. 273 
 
 existing, then the force exerted on that unit of electricity is 
 the force which we have to deal with in this case. If we 
 can-measure that force we can measure the density. This 
 method, however, is not generally applicable, because we 
 cannot usually place a charge outside a conductor without 
 disturbing the condition of the -conductor itself. The 
 generally applicable method is that by means of the carrier- 
 plane, which I spoke of before. 
 
 The next fundamental magnitude is electrical Potential. 
 This, in the case of electricity, corresponds to temperature 
 in the case of heat, or to the surface level in the case 
 of a liquid. The properties of any charged conductor 
 depend not only on the quantity of electricity it contains, 
 but also upon what we may call in general the quality of 
 the electricity, or the potential. Just so the properties of a 
 body containing heat depend not only on the quantity of 
 heat in the body, but upon the temperature of that heat, or 
 the temperature of the body as we call it. Potential is a 
 magnitude of which we cannot avoid the discussion in 
 speaking of electrical phenomena, for it corresponds to 
 properties which are not expressed by any other term. The 
 potential of a conductor, or rather the potentials of two 
 conductors, determine whether, when they are connected 
 together, there is any passage of electricity from one to the 
 other, and there is no other property which in general 
 determines this. It is easy to convince ourselves that whether 
 electricity passes from one conductor to another or not, does 
 not depend simply on the quantity which either of them 
 contains. Electricity does not always go from a conductor 
 which contains a large quantity to one which contains a small 
 quantity. Suppose this small ball and a larger sphere 
 brought together, the sphere being charged to begin with ; put 
 the ball in contact with the sphere, and then separate them. 
 They are now both charged, and the sphere has got more, 
 electricity than the ball, but if you put them in contact again 
 there is no further discharge of electricity from one to the 
 other. We have unequal quantities, but we have not, there- 
 fore, a passage of electricity from one to the other. Again, it 
 is not a question of the density of the charge, for we have 
 a greater density on this small ball than on the sphere; 
 still when we make the contact no electricity passes. A 
 still more conclusive case of the same kind is .this. Take an 
 
274 LECTURES TO SCIENCE TEACHERS. 
 
 elongated conductor such as a cylinder, or still better one 
 which is rounded at one end and pointed at the other. You 
 know that the electrical density at the extremities is greater 
 than at the intermediate parts, and if one end is rounded and 
 the other has a sharp point, the density at the point is very 
 much greater than the density at the round end ; still the 
 electricity does not flow from the point to the round end, or 
 from the ends to the middle. Take again such a case as 
 this. If we charge this jar inside and touch the inside with 
 a carrier-plane, we cannot detect any electrification ; if we 
 touch the outside we find it electrified. If you charge the 
 inside or outside of the jar, and then examine the inside, 
 you get no charge from that, but you do get a charge from 
 the outside, so that examining it in this way the inside 
 of the jar appears to be neutral. Wo connect the jar with 
 the electroscope in a different way by taking an insulated wire, 
 and bringing one end in contact with the outside of the jar ; 
 the leaves of the electroscope diverge, showing that the jar 
 is charged. Bring the end of the wire inside the jar, and the 
 leaves will likewise diverge. Thus, the inside of the jar 
 examined by means of a carrier-plane appears to be neutral, 
 but examined by means of a wire connected with an 
 electroscope, it appears to be charged just as much as 
 the outside. Again, although you cannot discharge the jar 
 by lading out electricity from the inside by a carrier, you 
 can discharge it by touching the outside with a carrier, then 
 uninsulating the carrier, touching the jar again, and repeating 
 the process over and over again. You can discharge the jar 
 in this way from the outside but not from the inside ; but 
 if you use a wire connected with the earth, whether you 
 touch the inside or touch the outside of the jar, you equally 
 discharge it. 
 
 There is another way in which we can show that the 
 inside of the jar although there is no electric density upon it, 
 and it appears neutral, when we examine it by means of a 
 Carrier it is not in the same condition as a space not sur- 
 rounded by electricity. If we put in an insulated ball which 
 is neutral to begin with, the jar being charged on the outside, 
 and, when the ball is inside, touch the ball with a wire in 
 connection with the earth and then take it out, we find the 
 ball charged, although there is no charge on the inside of 
 the jar. Thus, although the space within the jar does not 
 
ELECTRICAL MEASUREMENTS. 275 
 
 contain any electricity, it is not similar to a perfectly neutral 
 space. In fact the condition of the jar can only be expressed 
 by means of the term potential. If the jar is charged with 
 positive electricity the potential of the inside is higher than 
 the potential of the space outside. One result of this is that 
 positive electricity tends to pass away from any conductor 
 put inside the jar, and will pass away if there is a communi- 
 cation to allow it to do so. Electricity always tends to pass 
 from points of high potential, to points of lower potential 
 just as heat tends to pass from points of high temperature to 
 points of lower temperature. 
 
 The precise definition of potential at any point may be 
 stated in this way : Suppose we have a disc charged with a 
 unit of positive electricity, and we bring it to the point in 
 question from a place where no electric force acts, that is from 
 a great distance from any charged conductor. To make 
 sure that we begin far enough away, we start at an infinite 
 distance, and carry the unit of electricity up to the given 
 position. The total amount of work that we have done 
 in bringing the electricity to that position is the potential of 
 the place that we bring it to. If v/e bring it into a conductor 
 it is the potential of the conductor. The electrical potential 
 of any point is the total amount of work we must do to bring 
 to that point a unit of electricity from a position so far off 
 that no electric force acts at the point we bring it from. 
 Instruments for measuring potential and differences of potential 
 are called Electrometers, to which class of instruments two 
 subsequent lectures are to be devoted, so that I will not enter 
 further upon this question. 
 
 The next magnitude is the Capacity of conductors. Elec- 
 trical capacity is exactly comparable to the capacity for heat 
 of any body, or to capacity in its ordinary sense. The 
 capacity of a vessel for liquid is the quantity of liquid which 
 it can contain, only in the case of electricity the quantity 
 which a conductor can hold depends upon the means that 
 we employ for putting electricity into it. The capacity of a 
 conductor is perhaps better compared, not to the capacity of a 
 vessel for liquid, but to its capacity for a gas. The quantity 
 of air we can put into any space depends upon the pressure 
 of the air upon it. Without increasing the space we can put 
 twice as much air in if we double the pressure. So in the 
 case of a particular conductor, we can put twice as much 
 
 T 2 
 
276 LECTURES TO SCIENCE TEACHERS. 
 
 electricity into it if we double its potential. The electrical 
 capacity of the conductor is the quantity of electricity required 
 to produce unit change of potential. If the potential of the 
 conductor be zero before we give to it the charge, then the 
 quantity required to charge it to unit potential measures its 
 capacity. In general terms, if C stands for the capacity of a 
 conductor, and if the quantity Q of electricity given to that 
 conductor increases its potential by the amount V, then C = 
 
 ^., and ^ would be the quantity required to increase its 
 
 potential by unity. Or, to state the relation in another way, 
 the total quantity of electricity a conductor contains equals 
 the product of its capacity into the potential of the charge, 
 or Q = CY, just as the quantity of gas in a vessel might be 
 measured by the product of the volume into the pressure. 
 
 From this relation you see that if we know the quantity and 
 potential or their ratio, we know the capacity ; if we know 
 the capacity and potential, we know the quantity ; or again 
 if we know the capacity and the quantity we can determine 
 the potential, so that the measurement of these three things 
 cannot be separated from each other. The quantity of 
 electricity in a conductor can be measured by one or other 
 of the processes I have indicated, and its potential can 
 be measured by electrometers, and the ratio of these measures 
 the capacity. That is the general method of absolute 
 measurement. When we have merely to compare different 
 quantities, we may employ methods analogous to those 
 employed for measuring capacities for heat. For instance, to 
 determine the specific heat, or capacity for heat, of a substance 
 we put this substance at a known temperature into a known 
 mass of water at another known temperature, and observe 
 the changes of temperature of the two. The temperature of 
 the immersed body falls and that of the water rises, and they 
 come to equilibrium at some intermediate point. The same 
 quantity of heat is lost by one that is received by the other, 
 and the capacities for heat of the immersed body and of 
 the water are inversely proportional to the changes of 
 temperature which they respectively undergo. So in com- 
 paring the capacities of two electrical conductors : suppose 
 we have one of them charged to the potential V, and the 
 other at zero or uncharged, then, when we allow them to 
 come into contact, they both assume a common potential, 
 
ELECTRICAL MEASUREMENTS. 277 
 
 say v, intermediate between these two. One body has lost 
 the same quantity of electricity that the other has received. 
 The potential in. one case has fallen through V - v t and in 
 the other it has risen through + v and the capacities are 
 in the inverse ratio of the changes of potential caused by the 
 loss or gain of the same quantity of electricity ; so that if 
 the first body has the capacity C, and the second body the 
 capacity c, the capacities are to each other in the inverse 
 
 ratio of the changes of potential, or ^ ^-^-. That is the 
 
 < v 
 
 principle of a great number of methods of comparing the 
 
 capacities of different conductors. There is not enough time 
 to enter on the experiment in detail, or even on the principle 
 of other methods which might be given. If we have 
 the capacity of any one conductor measured in absolute 
 measure, the capacity of others can also be determined in 
 absolute measure by some process of comparison such as I 
 have indicated ; if we know the absolute value of one 
 thing, and can compare others with it, we know their 
 absolute value also. Upon the table are standard con- 
 densers or accumulators, as they are better called. They are 
 conductors of carefully ascertained capacity. They are of 
 different forms and different values. One is a very large one 
 formed practically by a combination of small ones. Here 
 again is an instrument devised by Sir William Thomson 
 which is a conductor of variable capacity ; its capacity can be 
 diminished by drawing out the cylinder, and increased by push- 
 ing it in, so that it can be adjusted so as to have the same 
 capacity as any particular conductor we have to measure and 
 then we can read off on the scale the capacity of the other one. 
 
 The Energy of charged conductors is proportional both to 
 the quantity and to the potential ; the absolute energy of a 
 charge of electricity is equal to half the product of the charge 
 into the potential, or ^ QV = E. That can be proved by an 
 apparatus like this. You discharge the electricity through a 
 fine wire stretched across the bulb of an air thermometer and 
 observe the amount of heat. This gives us a quantity which 
 is proportional to the energy of the charge. 
 
 I must be content with merely indicating these points ; and 
 in the lecture to follow on Wednesday next I shall have to 
 speak of some of the chief measurements connected with 
 electrical currents or dynamical electricity. 
 
ELECTRICAL MEASUREMENTS. 
 
 LECTURE II. 
 
 THE measurements connected with electrical currents 
 dynamical electricity are chiefly three. We have 1. 
 The measurement of the Strength of Currents, or, to express 
 it more shortly, the measurement of electric currents ; 
 2. The measurement of Electromotive Force ; and 3. The 
 measurement of the Resistance of Conductors. The measure- 
 ment of resistance or of conducting power comes to the 
 same thing, for the relation between these two properties 
 is a reciprocal one. If we know the conducting power of 
 any piece of wire or other conductor, we know that the 
 resistance is equal to 1 -^ the conducting power or the 
 conducting power is equal to 1 -f- the resistance ; so that 
 if the conducting power of a wire is one-half, its resistance 
 is two ; and if the resistance is one-half, the conducting 
 power is two, and so on. One is the reciprocal of the other, 
 so that whether we speak of the measurement of resistance 
 or of conducting power it is essentially the same thing. A 
 statement of the one is a statement in another form of the 
 other. 
 
 What I shall try to explain to-day, as far as time and 
 circumstances permit, is the methods by which absolute 
 measures of these three quantities can be obtained. 
 When I speak of the absolute measure of a current, I 
 mean a measurement which tells us, not that one current, 
 for instance, is twice or three times as strong as some 
 other current, but which tells us the actual quantity of 
 electricity conveyed by the current in unit of time. The 
 idea attached to the term strength of current in the case 
 of electricity is exactly comparable to the idea attached to 
 
ELECTRICAL MEASUREMENTS. 279 
 
 the corresponding term in the case of a current of anything 
 else. If we are speaking of a current of water, we mean 
 by the strength of a current the quantity of water 
 delivered in a given time, say a minute. Similarly 
 the quantity of electricity conveyed along a conductor 
 by a given current in a second or minute is the strength 
 of the current. Yery frequently, however, processes are 
 spoken of as measurements of the strength of currents 
 which really enable us merely to compare one current with 
 another. They are comparative measures, but not absolute 
 measures. We might have several water pipes delivering 
 water each at its own rate, and we might possibly ascertain 
 that one delivers three times as much per minute as another, 
 and another one sixteen times as much as the first, without 
 knowing what is the absolute quantity delivered by either 
 of them, and we have in electrical, as in all other measure- 
 ments, to distinguish between such comparative measure- 
 ments and absolute measurements which tell us, not only 
 whether a thing is greater or less than some other thing 
 of the same kind, but actually how great it is. And it is 
 the general principles of the methods by which absolute 
 measurements of the strength of currents can be obtained 
 that I shall have to speak of. 
 
 There are three chief properties or three effects pro- 
 duced by electric currents, any one of which might be 
 made the basis of a system of measurement. You all 
 know that when an electric current passes near a magnetic 
 needle it exerts force upon the needle, and usually dis- 
 places the needle from its ordinary position; a current 
 passing near a magnet deflects it, so that we may say that 
 an electric current produces electro magnetic effects ; 
 and we may use the term electromagnetic action of a 
 current to include all the mutual effects which take place 
 between currents and magnets. Upon this action of 
 electric currents then we might found a system of absolute 
 measurement. Then, again, one electric current passing 
 near another electric current exerts a force upon it, or 
 rather when two currents are near each other there is in 
 general a mutual force tending to displace them relatively 
 to each other. The forces exerted by electric currents 
 upon each other, usually spoken of as electro-dynamic 
 forces, might serve as the basis of a second system of 
 
280 LECTURES TO SCIENCE TEACHERS. 
 
 measurement. Again, an electric current passed through 
 a compound liquid almost always causes chemical 
 change a decomposition of the substance. The general 
 name given to processes of this kind is electrolysis the 
 electrical breaking up of the substance. And the electro- 
 lytic action of currents might be employed to furnish a 
 third method of measuring their strength. 
 
 In order to employ any one of these principles for the 
 measurement of currents, the conditions under which these 
 actions take place must be rendered as simple and definite 
 as possible. Without going into minute particulars I may 
 remind you to take the electro magnetic action first that 
 the force exerted by an electric current upon a magnet 
 depends partly on the strength of the current, partly upon 
 the relative positions of the magnet and of the conductor 
 in which the current passes. This I may almost say is self- 
 evident, that a change of relative position would change 
 the effect. One kind of change of relative position would 
 be a change of distance : the farther a current is from a 
 magnet, obviously the less it will act upon it ; but inde- 
 pendently of changes of distance, a change of direction 
 will in general cause a change of action ; so that, to express 
 the matter in general terms, we may say that the effect of 
 a current upon a magnet depends on the relative positions 
 of the two. It also depends upon the length of the con- 
 ductor which is in the neighbourhood of the magnet. That, 
 in a certain sense, is merely a repetition of what I said as 
 to position, and will be evident at once. If you take, for 
 instance, what I have here a magnet suspended on a pivot 
 at the centre of a graduated circle, with a circular con- 
 ductor formed by a coil of wire which would enable us 
 to pass a current many times round the magnet, and 
 terminated by wires by which it can be connected with 
 external conductors if those wires are carried far away 
 they may go to such a distance that the distant portion 
 of them does not exert any sensible influence on the magnet. 
 So that we have to consider the length of the conductor 
 near the magnet, not the total length of the conducting 
 wire. We might have a current sent through this coil 
 from a mile or 100 miles away, but the portion of the con- 
 ductor at that distance would obviously exert no sensible 
 effect. So then, the force exerted upon the magnet depends 
 
ELECTRICAL MEASUREMENTS. 281 
 
 upon the strength of the current, the relative positions of 
 the current and magnet, and the length of the conductor 
 which is within acting range of the magnet. The effect 
 also depends upon the strength of the magnet itself ; but 
 we may consider that we always deal with a magnet of 
 unit strength, and may therefore leave it out of con- 
 sideration now. The simplest rule to adopt to obtain a 
 standard of measurement founded upon this action, is 
 then to say that we will call that current a current 
 of unit strength, of which a unit length placed at 
 unit distance from any given point causes unit magnetic 
 force at that point. If you consider that statement you 
 will see at once some of the conditions that are necessary 
 in order to apply a system of measurement founded upon 
 the principle we are dealing with. If we are to have a 
 given length of a conductor, all of it at some definite 
 distance from a particular point, it follows at once that 
 the conductor must be bent into a circular arc. If we 
 have a line and every point 'of this line is to be at the same 
 distance from some particular point, the line must not be 
 straight, but it must be curved into the arc of a circle of 
 which this point is the centre. In order not to complicate 
 the matter by using unusual units, I will suppose we 
 take a foot as the unit of length. Take a wire a foot long 
 and bend it into the arc of a circle of one foot radius, so 
 that each point of it is at one foot distance from the centre. 
 Then suppose that we have an electric current flowing 
 along this wire ; this current will cause any magnet near it 
 to be acted upon by a force in one direction or the other, 
 and suppose that we have a magnet at the centre point. 
 If the magnetic force exerted at this particular point by the 
 current is the force one, say a force equal to the weight of 
 one grain, the strength of the current which flows along 
 the conductor is the standard of measurement it is repre- 
 sented by unity. And if the force exerted here is 2, 
 the strength of the current also is represented by 2 ; or 
 if the force at the centre is 100, the strength of the current 
 is 100. So that we take as the standard of measurement 
 of the strength of the current, the magnetic force exerted 
 at the centre of a circle of unit radius by unit length of 
 the current bent round the circumference of the circle. 
 Practically, this precise arrangement would not be con- 
 
282 LECTURES TO SCIENCE TEACHERS. 
 
 venient, and, therefore, instead of taking a wire, say one foot 
 long, and putting that partly round the circumference of a 
 circle of a foot radius we take a complete circumference ; that 
 is, if the radius of the circle is 1, the circumference 
 would be about 6*28 ; but the magnetic force at the centre 
 of the circle will be just proportional to the length of the 
 conductor. If we take one foot round a circle then unit 
 current would exert unit force ; but if we take six feet 
 round the circle then unit current in the circumference 
 would exert six times unit force; so we consider that 
 unit current is the current which, going once round a circle 
 of unit radius, exerts at the centre, not unit force, but 6 '28 
 units of force. 
 
 This principle is adopted in all such instruments as 
 these, where we have a current carried once round or several 
 times round a circle. If I send a current round this circle 
 I shall be able to cause a deflection of the magnet at the 
 centre. You will see from what I have said timt the 
 problem we have to deal with in measuring an electric 
 current by the method I have been speaking of, is to take 
 a measured length of the current at a measured distance 
 from a definite point, and to measure the magnetic force 
 exerted at that point. The easiest way of measuring the 
 magnetic force exerted at the centre of the circle is to 
 observe the deflection from the natural position which a 
 small magnet placed at the centre undergoes. If we know 
 the magnetic force exerted at any given point due to the 
 magnetic force of the earth, then it is easy to ascertain what 
 is the force exerted on the magnet when it is deflected 
 through a measured angle from the position it would take 
 if acted upon by the force of the earth only. We have a 
 little magnet at the centre of the circle, acted upon by the 
 earth, and it takes a definite position in consequence. If 
 we act upon it also by an electric current we in general 
 displace it from the position it would take if acted on by 
 the earth only, and if we know the force exerted upon the 
 magnet by the earth, and if we know also the extent to 
 which it is deflected by an electric current, then we can 
 measure the force exerted by the electric current. In that 
 way we get to know what the force is which the current 
 exerts ; by measuring the radius of the circle, and the 
 number of times which the wire goes round, we get the 
 
ELECTRICAL MEASUREMENTS. 283 
 
 length of the current which acts upon the magnet as well 
 as its distance from it; and in this way we have the 
 measurement complete. 
 
 The apparatus which may be used for applying a 
 measurement upon these principles may receive very 
 various forms. This is one of an essentially very simple 
 kind. It has the advantage that it can be employed to 
 measure currents of very different degrees of strength. 
 If we have a very strong current it may deflect a magnet 
 so much that we could not readily deduce the amount of 
 force exerted. If there is too great a force applied to any 
 particular instrument its indications are not so certain. 
 We have a familiar example of this in the case of a spring 
 balance. It will indicate, perhaps, very accurately, weights 
 up to a certain amount, but if you put on too great a weight 
 then the indication becomes uncertain. So any particular 
 measuring instrument acts best for quantities which lie 
 within a certain range ; and it generally requires a different 
 instrument or different adjustment of the same instrument 
 to measure quantities differing greatly from one another. 
 I may just point out the way in which this particular 
 instrument may be modified to suit stronger currents than 
 those which act upon it as now arranged ; that is by dis- 
 placing the circular conductor from the magnet. The 
 magnet is not at the centre of the circle, but each portion 
 of the wire, instead of surrounding the magnet, is on the 
 surface of a cone, of which the magnet is at the apex, and 
 the further we move it in this direction the smaller becomes 
 the effect of each portion of the wire, and therefore the 
 stronger the current required to give a definite deflection. 
 Here is another form of apparatus for the same purpose. 
 It is essentially the same instrument although smaller in 
 size and in various ways differently arranged. There is 
 a point at the centre for carrying a small magnet, a gradu- 
 ated circle to show how much the magnet is deflected, and 
 a vertical circle in which the current is to pass. Most of 
 you are aware that in using an instrument such as either 
 of these, it must be placed in a certain position ; that the 
 plane containing the circular conductor must coincide with 
 what is called the plane of the magnetic meridian. The 
 circle must in the first place be vertical, and it must also 
 coincide with the plane of the earth's magnetic force. 
 
284 LECTURES TO SCIENCE TEACHERS. 
 
 That in this room coincides pretty nearly with the length 
 of the room. The larger instrument is, therefore, im 
 properly placed, being across the room, so that it would 
 have to be turned round. The circle being properly set, 
 you know also that the strength of the current is in pro- 
 portion, not to the deflection of the magnet, but to the 
 tangent of that deflection; so that if the magnet is 
 deflected through any angle, the strength of the current 
 is proportional to the tangent of this angle. Instruments 
 such as these are therefore usually termed Tangent Gal- 
 vanometers ' } galvanometer being the name given to any 
 instrument which measures the strength of a current by 
 the electro-magnetic action of that current. 
 
 Here is an instrument, one of the first, I believe, of 
 the kind which was constructed ; it would not however, 
 easily give us an absolute measurement. There is a magnet 
 and a coil of wire surrounding it, so that when a current 
 passes through the coil the magnet is deflected ; but the 
 section of the coil which is wound on this elongated reel 
 is not circular, so that it would involve a very complicated 
 calculation in order to ascertain the mean distance of the 
 various parts of the current from the magnet. Such an 
 instrument could hardly be made to serve for absolute 
 measurement, but it will serve very well for comparative 
 measurements, and for that purpose it is so arranged that 
 the coil can turn independently of the magnet, and it is 
 used in this way : If we send a current through the coil 
 and that deflects the magnet, then the coil and the plate 
 which carries it are turned round until a fixed mark is 
 brought to coincide with an index attached to the magnet, 
 and the measurement depends on the extent to which this 
 circle is turned. It is an instrument which will give very 
 accurate comparative measurements, but not absolute 
 measurements. On the table there are various forms of 
 instruments for giving comparative measurements, and one 
 also for giving absolute measurements. There is a 
 magnet suspended inside this copper-box, and it is in 
 the form of a magnetized steel ring. The thick mass of 
 copper which surrounds it is for the purpose of making 
 it come rapidly to rest when displaced. Probably you 
 know that when a magnet swings in the neighbourhood of 
 a large mass of copper or other conducting material, its 
 
ELECTRICAL MEASUREMENTS. 285 
 
 movement is damped, just as if it were swinging in oil or 
 some resisting fluid. Here is a circular plate with a 
 groove cut in it, in which a conducting wire can be placed, 
 and the diameter of the circle can be accurately measured. 
 The distance of the circle from the magnet is also 
 accurately measured by means of a divided scale. Here 
 we have again the means of obtaining a conductor of 
 accurately known length in an accurately known position 
 relatively to the magnet, and the deflection of the magnet 
 can be observed. The amount of deflection would be 
 ascertained in this case, not by a graduated circle, but 
 by means of a mirror which is attached to the magnet and 
 turns with it, and the deflection of the mirror is obtained by 
 means of a telescope. There is another apparatus here on 
 the same principle. If we place a divided scale opposite 
 the mirror, the divisions of the scale are reflected from the 
 mirror back into the telescope which is placed opposite. 
 In this way an extremely small deflection of the mirror, and, 
 therefore, of the magnet, can be detected. The method 
 of reading these two instruments is the same, but the first 
 is an absolute instrument and the second is merely a com- 
 parative apparatus. It is a very convenient form of its 
 kind. There are three pairs of coils belonging to it, one a 
 pair of long coils of thin wire, and the other pairs being 
 shorter and of thicker wire. There is the same arrange- 
 ment as in the first I showed you for varying the distance 
 between the conductor and the magnet. 
 
 Here again are other forms of galvanometers, but still 
 comparative instruments only. These are different forms 
 of Sir William Thomson's Reflecting Galvanometer, 
 perhaps one of the most delicate instruments of the kind 
 ever constructed to indicate the passage of an exceedingly 
 small current. It will give us the comparative measure- 
 ment of very weak currents, but not an absolute measure. 
 
 Here, again, is an apparatus, constructed by Professor 
 Guthrie, which will also give us comparative measurements. 
 It depends upon the mutual attraction or repulsion of little 
 bits of iron magnetized by the passage of a current. 
 
 So much for the various ways in which electro-magnetic 
 action can be made to serve as a basis for a system of 
 measurement. It is much the most frequently employed 
 principle, but we can also employ electro-dynamic action 
 
286 LECTURES TO SCIENCE TEACHERS. 
 
 the force exerted by one current on another, or by one 
 portion of a current on another portion of the same 
 current. There are on the table two instruments which 
 act on this principle. Professor Guthrie's may be regarded 
 as one of this class. The same current passes round two 
 pairs of small pieces of iron, and it is the action of one 
 part of the current on another which really gives us 
 the indication. Here, again, is an apparatus where the 
 electro-dynamic action of the current is applied more 
 simply, and without the intervention of the bits of iron. 
 We have a circular coil of wire, to begin with, through 
 which the current passes ; then instead of suspending a 
 magnet at the centre of the coil, there is another coil 
 suspended inside, and the two coils are so set that when 
 no current is passing, the planes of the separate portions of 
 the wire are at right angles to each other. The wire in 
 the fixed coil is in a plane parallel to the length of the 
 table, but in the suspended coil the wire is wound almost 
 at right angles to that direction. When it is properly 
 adjusted, the directions of winding the two coils are at right 
 angles to each other, but when a current passes through 
 both of them there is a tendency in the two coils to set 
 parallel to each other. If they are at right angles to 
 begin with, a force which would tend to set them parallel, 
 will cause a deflection, and the amount of this deflection, 
 combined with a knowledge of the force resisting the 
 displacement, and of the relative lengths and positions of 
 the two parts of the conductors which act upon each other, 
 gives us again the means of ascertaining the strength of the 
 current. When we employ this principle the statement 
 which involves the definition of the unit current may be 
 put in this way : Unit current placed around the circum- 
 ference of a circle of unit area exerts upon an equal current, 
 enclosing an equal area in a plane at right angles to the 
 first, a couple whose moment multiplied by the cube of the 
 distance between the areas surrounded by the currents 
 equals unity when the distance is very great. In order to 
 make the statement of the mechanical force simple, you 
 must imagine that the two currents are at a very great 
 distance from each other, as compared to the size of 
 the area surrounded by either of them; the force ex- 
 erted by one upon the other diminishes as the distance 
 
ELECTRICAL MEASUREMENTS. 267 
 
 increases, but, if you multiply the force by the cube of 
 the distance, then the product approaches more and more 
 nearly to a constant value as the distance becomes greater, 
 and if the current is one of unit strength, and the two 
 areas are equal to unity, the constant value to which this 
 product approaches as the distance increases is unity. But, 
 as you will see, the considerations involved are rather 
 complex, and the further discussion of this subject would 
 take much more time than we can spare. 
 
 Then another action produced by currents which might 
 be adopted, and often is adopted, for measuring their 
 strength, is the amount of chemical change which they can 
 produce in a given time ; or, to express it more concisely, 
 the strength of a current can be measured by the rate at 
 which it can produce chemical change in a body through 
 which it is passed by the rate, for instance, at which it 
 can decompose water or any other chemical compound. 
 Adopting this system we may take as the unit current one 
 of such a strength that it will decompose unit mass of a 
 standard substance say water in unit of time say a 
 second. We may define unit current as a current which 
 will decompose a grain of water in a second, taking a 
 grain as the unit of mass, water as the standard substance, 
 and a second as the unit of time. But if we are to adopt 
 this definition, the unit current obtained in this way would 
 not agree with the unit current derived from the electro- 
 magnetic action. An electro-magnetic unit current (re- 
 ferred to centimetres, grammes, and seconds), if used to 
 decompose water, will liberate '0001052 gramme of 
 hydrogen per second, or it would decompose nine times 
 that quantity of water. This gives us the ratio between 
 the electro-magnetic unit current and this electrolytic 
 unit. The apparatus employed for carrying out this kind 
 of measurement is of various kinds. They are called Volta- 
 meters ; many forms are familiar to all who have dealt at 
 all with electrical processes, and I will, therefore, merely 
 speak of one form which may be new to most if not all of 
 you a very beautiful instrument, constructed by Professor 
 Lenz, of St. Petersburg, which Baron von Wrangel has 
 kindly explained to me. It is an apparatus in which a 
 basic mercurous nitrate is decomposed. The current- is 
 passed through two little glass vessels containing mercury, 
 
288 LECTURES TO SCIENCE TEACHERS. 
 
 communicating with each other by a narrow tube, the 
 mercury in both vessels being covered with a solution of 
 nitrate. The current passing causes more mercury to be 
 dissolved in one vessel, and more mercury to be separated 
 from the solution in the other, so that there is a continual 
 carrying of mercury from one vessel to the other, by the 
 action of the current. The rate of increase in the lower 
 vessel, or, in other words, the quantity carried over in 
 a given time measures the strength of the current. The 
 apparatus is so arranged as to give the means of observing 
 with very great accuracy the rate of increase in the 
 quantity of mercury. There is a micrometer screw, by 
 means of which the surface of the mercury can be lowered, 
 and read off exactly against a mark, and then, after the 
 experiment, the same measurement is made again, and the 
 difference between the two readings of the screw indicates 
 the increase in the volume of mercury. It is said that in 
 the course of a minute or two a sufficient quantity of 
 mercury is carried over to give a measurable increase, 
 and that the measurement can be made with very great 
 accuracy indeed. The essential electro-chemical principles 
 involved are the same as in the ordinary voltameter, but 
 there is a different mechanical arrangement for observing 
 the amount of effect produced. 
 
 I have only time to state very briefly the general principles 
 of the methods available for measuring Electromotive Force. 
 In the most usual cases electric currents are produced 
 either by chemical action or by the relative motion of 
 magnets and conductors. In an ordinary galvanic battery 
 we have chemical action taking place as long as the current 
 passes, the quantity of the current which traverses the 
 circuit being proportional to the quantity of chemical action 
 which takes place. In apparatus, such as this on the table, 
 we have currents produced by the motion of magnets and 
 conducting-wires relatively to each other. An absolute 
 measure of electro-motive force may be derived from a study 
 of the conditions under which currents are produced in 
 either of these cases. 
 
 Thus, first, if we can determine the amount of energy 
 expended in a galvanic battery when a unit of electricity 
 traverses the circuit, we have really calculated the electro- 
 motive force. Now the quantity of heat corresponding to 
 
ELECTRICAL MEASUREMENTS. 289 
 
 the chemical action has in very many cases been determined, 
 and when we know the quantity of heat, we have only to 
 multiply this by the mechanical equivalent of heat, and we 
 get the electro motive force ; for the electro-motive force is 
 the work done, or the amount of energy expended in the 
 circuit when unit of electricity traverses the conductor. 
 From the determination of the heat of chemical action, and 
 the mechanical equivalent of heat we have the data for 
 calculating the electro-motive force due to any given 
 chemical process, such as takes place in a galvanic 
 battery. 
 
 Again, if the magnetic forces acting upon a conductor 
 which we move in a field of magnetic force are definitely . 
 known, and the motion of the conductor relatively to the 
 magnetic field is definitively known, we can again calculate 
 the quantity of work done in maintaining the motion, and 
 hence also the electro-motive force produced. Thus the 
 electro-motive force may be theoretically ascertained either 
 from the energy of the chemical action required to produce 
 the current, or from the work done in maintaining the 
 motion of conductors in the neighbourhood of magnets ; 
 though I cannot enter into the details in either case. 
 
 When we have got a measurement of the current and of the 
 electro-motive force, the measurement of Resistance follows 
 from a comparison of the two. The electro-motive force 
 acting in any circuit divided by the strength of the current 
 which that electro-motive force produces, measures the 
 resistance of the circuit. That is the general principle of 
 all absolute methods of measuring the resistance of circuits. 
 There is here an apparatus by which such a comparison 
 can be made. This is a circular conductor which can rotate 
 about a vertical axis, and as it rotates a current is produced 
 in it by the earth's magnetic force. By suspending a little 
 magnet inside this little box at the centre of the circle the 
 strength of the current produced in the wire can be ascer- 
 tained by the deflection of the magnet. The deflection of 
 the magnet combined with the speed of the rotation and 
 the dimensions of the coil gives us the absolute resistance 
 of the conductor. This is the actual apparatus employed 
 in a very elaborate series of measurements by a Com- 
 mittee of the British Association, of which Professor Clerk 
 k Maxwell was the most active member. It was employed 
 ' 
 
290 LECTURES TO SCIENCE TEACHERS. 
 
 in determining the so-called British Association Unit of 
 Resistance. Here is the coil with which the resistance of 
 the revolving apparatus was compared periodically during 
 the course of the experiments ; and here are multiples of 
 the standard so established, by ten, one hundred, and a 
 thousand. 
 
 END OF VOL. i. 
 
 LONDON: R. CLAY, SONS, AND TAYLOR. 
 
MACMILLAN'S 
 
 SCIENCE PRIMEB8. 
 
 UNDER THE JOINT EDITORSHIP OF 
 
 PROFESSORS HUXLEY, ROSCOE, AND 
 
 BALFOUR STEWART. 
 CHEMISTRY. By H. E. ROSCOE, F.R.S., Professor of 
 
 Chemistry in Owens College, Manchester. 18mo. Illustrated. 1*. With 
 Questions. 
 
 PHYSICS. By BALFOUE STEWART, F.R.S., Professor of 
 
 Natural Philosophy in Owens College, Manchester. 18mo. Illustrated, la. 
 With Questions. 
 
 \ PHYSICAL GEOGRAPHY. By A. GEIKIB, F.R.S., 
 
 Murchison Professor of Geology and Mineralogy at Edinburgh. 18rao. 
 Illustrated. Is. With Questions. 
 
 , GEOLOGY. By Professor GEIKIE, F.R.S. 18mo. With 
 
 numerous Illustrations. Is. 
 
 PHYSIOLOGY. By MICHAEL FOSTER, M.D., F.R.S. 
 
 18mo. With numerous Illustrations. Is. 
 
 /ASTRONOMY. By J. NORMAN LOCKYER, F.R.S.' 18mo. 
 
 With numerous Illustrations. Is. 
 
 BOTANY. By Sir J. D. HOOKER, K.C.S.I., C.B., 
 
 President of the Royal Society. 18mo. Illustrated. Is. 
 
 LOGIC. By Professor STANLEY JEVONS, F.R. S. 18mo. Is. 
 POLITICAL ECONOMY. By Professor JEVONS, F.R.S. 
 
 18mo. Is. . : 
 
 INTRODUCTORY. By Professor HUXLEY, F.R.S. 
 
 [Preparing, with others. 
 
 MANCHESTER SCIENCE LECTURES. 
 
 EIGHTH SERIES, 1876-77. 
 
 Crown 8vo. With Illustrations, price Qd. each. Also complete in one volume, 
 cloth, 2s. 
 
 WHAT THE EARTH IS COMPOSED OF. By Pro- 
 fessor ROSCOE, F.R.S. 
 
 THE SUCCESSION OF LIFE ON THE EARTH. By 
 
 Professor W. C. WILLIAMSON. 
 
 WHY THE EARTH'S CHEMISTRY IS AS IT IS. 
 
 By J. NORMAN LOCKYER, F.R.S. 
 
 MACMILLAN AND CO., LONDON. 
 
MACMILLAN'S 
 SCIENCE CLASS-BOOKS. 
 
 ANATOMY. ELEMENTARY LESSONS IN. By ST. 
 
 GEORGE MIVART, F.R.S. With Illustrations. ISmo. 6s. 6d. 
 
 ASTRONOMY. POPULAR. By Sir G. P. AIRY, C.B., 
 
 Astronomer-Royal. With Illustrations. New Edition. 18mo. 4s. 6d. 
 
 ^ASTRONOMY. ELEMENTARY LESSONS IN. By 
 
 J. N. LOCKYER, F.R.S. With Coloured Diagram of the Spectra of the 
 Sun, Stars, Nebulae, and other Illustrations. New Edition. l8mo. 5s. Gd. 
 QUESTIONS ON THE SAME. Is. 6d. 
 
 BOTANY. LESSONS IN ELEMENTARY. With Illus- 
 trations. By Professor OLIVER, F.R.S. New Edition. 18mo. 4s. 6d. 
 
 ^CHEMISTRY LESSONS IN ELEMENTARY. By 
 
 Professor ROSCOE, F.R.S. With Illustrations and Chromo-lithograph. New 
 Edition. 18ino. 4s. 6d. 
 
 CHEMISTRY. OWENS COLLEGE JUNIOR COURSE 
 
 OF PRACTICAL CHEMISTRY. By F. JONES. With Preface by Pro- 
 fessor KOSCOK. New Edition. 18mo. 2s. 6d. 
 
 V LOGIC. ELEMENTARY LESSONS IN ; DEDUCTIVE 
 
 AND INDUCTIVE. By Professor JtiVONS, F.R.S. With Questions, Exam- 
 ples, and Vocabulary. New Edition. ISmo. 3s. (jd. 
 
 PHYSIOLOGY. LESSONS IN ELEMENTARY. With 
 
 numerous Illustrations. By Professor HUXLEY, F.R.S. New Edition. 18mo. 
 4s. 6d. 
 QUESTIONS ON THE SAME. Is. 6d. 
 
 VpHYSICS. LESSONS IN ELEMENTARY. By Pro- 
 fessor BALFOUR STEWART, F.R.S. With Coloured Diagram and numerous 
 Illustrations. New Edition. 18mo. 4s. 6d. 
 
 POLITICAL ECONOMY FOR BEGINNERS. By 
 
 MILLICENT GARRETT FAWCETT. With Questions. New Edition. 18mo. 
 2s. tid 
 
 STEAM : AN ELEMENTARY TREATISE. By JOHN 
 
 PERRY, C.E. With numerous Illustrations, Examples, and Exercises. 18mo. 
 4s. 6d. 
 
 NATURAL PHILOSOPHY FOR BEGINNERS. By 
 
 I. TODHUNTER, M.A., F.R.S. Part I. Properties of Solid and Fluid Bodies. 
 18mo. 3s. 6d. Part II. Sound, Light, and Heat. 18mo. 3s. Qd. 
 
 PHYSICAL GEOGRAPHY. ELEMENTARY LES - 
 
 SONS IN. By Professor ARCHIBALD GEIK1E, F.R.S. With numerous 
 Illustrations. Fcap. 8vo. 4s. tid. 
 QUESTIONS. 1. 6d. 
 
 MACMILLAN AND CO., LONDON. 
 
BEDFORD STREET, COVENT GARDEN, LONDON, 
 September 1878. 
 
 MACMILLAN & Co.'s CATALOGUE of WORKS 
 in MATHEMATICS and PHYSICAL SCIENCE; 
 including PURE and APPLIED MATHE- 
 MATICS; PHYSICS, ASTRONOMY, GEOLOGY, 
 CHEMISTRY, ZOOLOGY, BOTANY; and of 
 WORKS in MENTAL and MORAL PHILOSOPHY 
 and Allied Siibjects. 
 
 MATHEMATICS. 
 
 Airy. -Works by Sir G. B. AIRY, K.C.B., Astronomer Royal.- 
 
 ELEMENTARY TREATISE ON PARTIAL DIFFERENTIAL 
 EQUATIONS. Designed for the Use of Students in the Univer 
 sities. With Diagrams. New Edition. Crown 8vo. 51. 6d. 
 
 ON THE ALGEBRAICAL AND NUMERICAL THEORY OF 
 ERRORS OF OBSERVATIONS AND THE COMBINA- 
 TION OF OBSERVATIONS. Second Edition. Crown 8vo. 
 6s. 6d. 
 
 UNDULATORY THEORY OF OPTICS. Designed for the Use of 
 Students in the University. New Edition. Crown 8vo. 6s. 6d. 
 
 ON SOUND AND ATMOSPHERIC VIBRATIONS. With 
 the Mathematical Elements of Music. Designed for the Use of 
 Students of the University. Second Edition, revised and enlarged. 
 Crown 8vo. gs. 
 
 A TREATISE ON MAGNETISM. Designed for the Use of 
 Students in the University. Crown 8vo. 9-r. 6d. 
 
 Ball (R. S., A.M.) EXPERIMENTAL MECHANICS. A 
 
 Course of Lectures delivered at the Royal College of Science for 
 Ireland. By ROBERT STAWELL BALL, A.M., Professor of Applied 
 Mathematics and Mechanics in the Royal College of Science for 
 Ireland (Science and Art Department). Royal 8vo. ids. 
 " We have not met "with any book of the sort in English. It eluci- 
 dates instructively the methods of a teacher of the very highest 
 rank. We most cordially recommend it to all our readers" 
 Mechanics' Magazine. 
 s,ooo. 8. 78. A 
 
SCIENTIFIC CATALOGUE. 
 
 Kelland and Tait. AN INTRODUCTION TO QUATER- 
 NIONS. With numerous Examples. By P. KELLAND, M.A., 
 F.R.S., and P. G. TAIT, M.A., Professors in the department of 
 Mathematics in the University of Edinburgh. Crown 8vo. Js. 6d. 
 
 Kempe. HOW TO DRAW A STRAIGHT LINE. A Lecture 
 on Linkages. By A.B. KEMPE, B.A. Illustrated. Crown 8vo. is.bd. 
 
 Merriman. ELEMENTS OF THE METHOD OF LEAST 
 SQUARES. By MANSFIELD MERRIMAN, Professor of Civil and 
 Mechanical Engineering, Lehigh University, Bethlehem, Penn., 
 U.S.A. Crown 8vo. Js. 6d. 
 
 Morgan. A COLLECTION OF PROBLEMS AND EXAM- 
 PLES IN MATHEMATICS. With Answers. By H. A. 
 MORGAN, M.A., Sadlerian and Mathematical Lecturer of Jesus 
 College, Cambridge. Crown 8vo. cloth. 6s. 6d. 
 
 Newton's Principia. 4 to. cloth. 31*. 6d. 
 
 It is a sufficient guarantee of the reliability of this complete edition of 
 Newton 's Principia that it has been printed for and under the care 
 of Professor Sir William Thomson and Professor Blackburn, of 
 Glasgow University. 
 
 Parkinson. A TREATISE ON OPTICS. By S. PARKIN- 
 SON, D.D., F.R.S., Fellow and Tutor of St. John's College, 
 Cambridge. Third Edition, revised and enlarged. Crown 8vo. 
 cloth. los. 6d. 
 
 Phear. ELEMENTARY HYDROSTATICS, with Numerous 
 
 Examples. ByJ. B. PHEAR, M.A., Fellow and late Assistant Tutor 
 of Clare ColL Cambridge. Fourth Edition. Cr. 8vo. cloth. $s. 6d. 
 
 Pirrie. LESSONS ON RIGID DYNAMICS. By the Rev. G. 
 PIRRIE, M. A. , Fellow and Tutor of Queen's College, Cambridge. 
 Crown 8vo. 6s. 
 
 Puckle. AN ELEMENTARY TREATISE ON CONIC SEC- 
 TIONS AND ALGEBRAIC GEOMETRY. With numerous 
 Examples and Hints for their Solution. By G. HALE PUCKLE, 
 M. A. Fourth Edition, enlarged. Crown 8vo. "js. 6d. 
 
 Rayleigh. THE THEORY OF SOUND. By LORD RAYLEIGH, 
 F.R.S., formerly Fellow of Trinity College, Cambridge. 8vo. 
 Vol. 1. I2J-. 6d. ; Vol. II. 12s. 6d. [Vol. III. in preparation. 
 
 Reuleaux. THE KINEMATICS OF MACHINERY. Out- 
 lines of a Theory of Machines. By Professor F. REULEAUX. 
 Translated and edited by A. B. W. KENNEDY, C.E., Professor of 
 Civil and Mechanical Engineering, University College, London, 
 With 450 Illustrations. Royal 8vo. 2QS. 
 
MATHEMATICS. 5 
 
 Routh. Works by EDWARD JOHN ROUTH, M.A., F.R.S., late 
 Fellow and Assistant Tutor of St. Peter's College, Cambridge ; 
 Examiner- in the University of London : 
 
 AN ELEMENTARY TREATISE ON THE DYNAMICS OF 
 THE SYSTEM OF RIGID BODIES. With numerous 
 Examples. Third Edition, enlarged. 8vo. 2U. 
 
 STABILITY OF A GIVEN STATE OF MOTION, PARTI- 
 CULARLY STEADY MOTION. The Adams' Prize Essay for 
 1877. 8vo. 8>. 6<t. 
 
 Tait and Steele. DYNAMICS OF A PARTICLE. With 
 
 numerous Examples. By Professor TAIT and Mr. STEELE. Fourth 
 Edition, revised. Crown 8vo. 12s. 
 
 Thomson. PAPERS ON ELECTROSTATICS AND MAG- 
 NETISM. By Professor SIR WILLIAM THOMSON, F.R.S. 
 8vo. i&y. 
 
 Todhunter. Works by I. TODHUNTER, M.A., F.R.S., of 
 
 St. John's College, Cambridge : 
 
 "Mr. Todhunter is chiefly known to students of mathematics as the 
 author of a series of admirable mathematical text-books, which 
 possess the rare qualities of being clear in style and absolutely free 
 from mistakes, typographical or other ." Saturday Review. 
 
 A TREATISE ON SPHERICAL TRIGONOMETRY. Third 
 
 Edition, enlarged. Crown 8vo. cloth. 4^. 6d. 
 PLANE CO-ORDINATE GEOMETRY, as applied to the Straight 
 
 Line and the Conic Sections. With numerous Examples. Fifth 
 
 Edition. Crown 8vo. 7s. 6d. 
 A TREATISE ON THE DIFFERENTIAL CALCULUS. 
 
 With numerous Examples. Seventh Edition. Crown 8vo. los. 6d. 
 A TREATISE ON THE INTEGRAL CALCULUS AND ITS 
 
 APPLICATIONS. With numerous Examples. Fourth Edition, 
 
 revised and enlarged. Crown 8vo. cloth, los. 6d. 
 EXAMPLES OF ANALYTICAL GEOMETRY OF THREE 
 
 DIMENSIONS. Third Edition, revised. Crown 8vo. cloth. 4^. 
 A TREATISE ON ANALYTICAL STATICS. With numerous 
 
 Examples. Third Edition, revised and enlarged. Crown 8vo. 
 
 cloth, icxr. 6d. 
 A HISTORY OF THE MATHEMATICAL THEORY OF 
 
 PROBABILITY, from the Time of Pascal to that of Laplace. 
 
 8vo. i8j. 
 RESEARCHES IN THE CALCULUS OF VARIATIONS, 
 
 Principally on the Theory of Discontinuous Solutions : An Essay 
 
 to which the Adams' Prize was awarded in the University of 
 
 Cambridge in 1871. 8vo. 6s. 
 
SCIENTIFIC CATALOGUE. ' 
 
 Todhunter continued. 
 
 A HISTORY OF THE MATHEMATICAL THEORIES OF 
 ATTRACTION, and the Figure of the Earth, from the time of 
 Newton to that of Laplace. Two vols. 8vo. 245-. 
 "Probably no man in England is so qualified to do justice to the 
 theme as Mr. Todhunter. To all mathematicians these volumes 
 "will be detply interesting, and to all succeeding investigators^ of the 
 highest practical utility. " Athenaeum. 
 
 AN ELEMENTARY TREATISE ON LAPLACE'S, LAME'S, 
 AND BESSEL'S FUNCTIONS. Crown 8vo. lor. 6d. 
 
 Wilson (W. P.) A TREATISE ON DYNAMICS. By 
 W. V. WILSON, M.A., Fellow of St. John's College, Cambridge, 
 and Professor of Mathematics in Queen's College, Belfast. 8vo. 
 gs. 6d. 
 
 Wolstenholme. A BOOK OF MATHEMATICAL 
 
 PROBLEMS, on Subjects included in the Cambridge Course. 
 By JOSEPH WOLSTENHOLME, Fellow of Christ's College, some 
 time Fellow of St. John's College, and lately Lecturer in Mathe- 
 matics at Christ's College. Crown 8vo. cloth. &s. 6d. 
 
 Young. SIMPLE PRACTICAL METHODS OF CALCU- 
 LATING STRAINS ON GIRDERS, ARCHES, AND 
 TRUSSES. With a Supplementary Essay on Economy in suspen- 
 sion Bridges. By E. W. YOUNG, Associate of King's College, 
 London, and Member of the Institution of Civil Engineers. Svo. 
 7-r. 6d. 
 
PHYSICAL SCIENCE. 
 
 PHYSICAL SCIENCE. 
 
 Airy (G. B.) POPULAR ASTRONOMY. With Illustrations. 
 By Sir G. B. AIRY, K.C.B., Astronomer Royal. New Edition. 
 i8mo. cloth. 4J. 6d. 
 
 Bastian. Works by H. CHARLTON BASTIAN, M.D., F.R.S., 
 Professor of Pathological Anatomy in University College, London, 
 c. : 
 
 THE BEGINNINGS OF LIFE : Being some Account of the Nature, 
 
 Modes of Origin, and Transformations of Lower Organisms. In 
 
 Two Volumes. With upwards of 100 Illustrations. Crown 8vo. 28.9. 
 
 "// is a book that cannot be ignored, and must inevitably lead to 
 
 renewed discussions and repeated observations, and through these to 
 
 the establishment of truth " A. R. Wallace in Nature. 
 
 EVOLUTION AND THE ORIGIN OF LIFE. Crown 8vo. 
 6j. 6d. 
 
 " Abounds in information of interest to the student of biological 
 science"- -Daily News. 
 
 Blake. ASTRONOMICAL MYTHS. Based on Flammarion's 
 "The Heavens." By John F. BLAKE. With numerous Illustra- 
 tions. Crown 8vo. ys. 
 
 Blanford (H. F.) RUDIMENTS OF PHYSICAL GEO- 
 GRAPHY FOR THE USE OF INDIAN SCHOOLS. By 
 H. F. BLANFORD, F.G.S. With numerous Illustrations and 
 Glossary of Technical Terms employed. New Edition. Globe 
 8vo. 2s. 6d. 
 
 Blanford (W. T.) GEOLOGY AND ZOOLOGY OF 
 ABYSSINIA. By W. T. BLANFORD. 8vo. 21*. 
 With Coloured Illustrations and Geological Map. " The result of 
 his labours" the Academy says, " is an important contribution 
 to the natural history of the country" 
 
 Bosanquet AN ELEMENTARY TREATISE ON MUSICAL 
 INTERVALS AND TEMPERAMENT. With an Account of 
 an Enharmonic Harmonium exhibited in the Loan Collection of 
 Scientific Instruments, South Kensington, 1876 ; also of an Enhar- 
 monic Organ exhibited to the Musical Association of London, 
 May, 1875. By R. H. Bosanquet, Fellow of St. John's College, 
 Oxford. 8vo. 6s. 
 
SCIENTIFIC CATALOGUE. 
 
 Cooke (Josiah P., Jun.) FIRST PRINCIPLES OF 
 CHEMICAL PHILOSOPHY. By JOSIAH P. COOKE, Jun., 
 Ervine Professor of Chemistry and Mineralogy in Harvard College. 
 Third Edition, revised and corrected. Crown 8vo. 12s. 
 
 Cooke (M. C.) HANDBOOK OF BRITISH FUNGI, 
 with full descriptions of all the Species, and Illustrations of the 
 Genera. By M. C. COOKE, M.A. Two vols. crown 8vo. 24^. 
 " Will maintain its place as the standard English book, on the 
 subject of which it treats ; for many years to come.''' 1 Standard. 
 
 Dawkins, CAVE-HUNTING: Researches on the Evidence of 
 Caves respecting the Early Inhabitants of Europe. By W. BOYD 
 DAWKINS, F.R.S., c., Lecturer in Geology at Owens College, 
 Manchester. With Coloured Plate and Woodcuts. 8vo. zis. 
 *' The mass of information he has brought together, with the judicious 
 use he has made of his materials, will be found to invest his book 
 with much of new and singular value." Saturday Review. 
 
 Dawson (J. W.) ACADIAN GEOLOGY. The Geologic 
 
 Structure, Organic Remains, and Mineral Resources of Nova 
 Scotia, New Brunswick, and Prince Edward Island. By JOHN 
 WILLIAM DAWSON, M.A., LL.D., F.R.S., F.G.S., Principal and 
 Vice- Chancellor of M'Gill College and University, Montreal, &c. 
 With a Geological Map and numerous Illustrations. Third Edition, 
 with Supplement. 8vo. 2is. Supplement, separately, 2s. 6d. 
 " The book will doubtless fnd a place in the library, not only of 
 the scientific geologist, but also of all who are desirous of the in- 
 dustrial progress and commercial prosperity of the Acadian pro- 
 vinces" Mining Journal. 
 
 Fleischer. A SYSTEM OF VOLUMETRIC ANALYSIS. 
 By Dr. E. FLEISCHER. Translated from the Second German 
 Edition by M. M. Pattison Muir, with Notes and Additions. 
 Illustrated. Crown 8vo. "js. 6d. 
 
 Forbes, THE TRANSIT OF VENUS. By GEORGE FORBES, 
 
 B.A., Professor of Natural Philosophy in the Andersonian Univer- 
 sity of Glasgow. With numerous Illustrations. Crown 8vo. 3-r. 6d. 
 " Professor Forbes has done his work admirably." Popular Science 
 Review. "A compact sketch of the whole matter in alt its as- 
 pects." Saturday Review. 
 
 Foster and Balfour. ELEMENTS OF EMBRYOLOGY 
 
 By MICHAEL FOSTER, M.D., F.R.S., andF. M. BALFOUR, M.A., 
 Fellow of Trinity College, Cambridge. With numerous Illustra- 
 tions. Part I. Crown 8vo. "js. 6d. 
 
 Galton. Works by FRANCIS GALTON, F.R.S. : 
 
 METEOROGRAPHICA, or Methods of Mapping the Weather. 
 Illustrated by up wards of 600 Printed Lithographic Diagram . 410. 9^. 
 
PHYSICAL SCIENCE, 
 
 Gait O n continued. 
 
 HEREDITARY GENIUS : An Inquiry into its Laws and Con- 
 sequences. Demy 8vo. 12s. 
 
 The Times calls it "a most able and mest interesting book;" and 
 Mr. Darwin, in his ll Descent of Man" (vol. \. p. Ill), says, " We 
 knoT.v, through the admirable labours of Mr. Galton, that Genius 
 tends to be inherited. " 
 
 ENGLISH MEN OF SCIENCE; THEIR NATURE AND 
 NURTURE. 8vo. Ss. 6d. 
 " The book is certainly one of very great interest." Nature. 
 
 Geikie. ELEMENTARY LESSONS IN PHYSICAL GEO- 
 GRAPHY. With numerous Illustrations. By ARCHIBALD 
 GEIKIE, LL.D., F.R S., Murchison Professor of Geology and 
 Mineralogy at Edinburgh. Fcap. 8vo. 4^. 6d. Questions, Is. bd. 
 
 Gordon. AN ELEMENTARY BOOK ON HEAT. Byj. E. 
 
 H. GORDON, B.A., Gonville and Caius College, Cambridge. 
 Crown 8vo. 2s. 
 
 Guillemin. THE FORCES OF NATURE.- A Popular intro- 
 duction to the Study of Physical Phenomena. By AMEDEE 
 GUILLEMIN. Translated from the French by MRS. NORMAN 
 LOCKYER ; and Edited, with Additions and Notes, by J. NORMAN 
 LOCKYER, F.R.S. Illustrated by Coloured Plates, and 455 Wood- 
 cuts. Third and cheaper Edition. Royal 8 vo. 2is. 
 " Translator and Editor have done justice to their trust. The 
 text has all the force and flow of original writing, combining 
 faithfulness to the author s meaning with purity and independence 
 in regard to idiom ; while the historical precision and accuracy 
 pervading the work throughout, speak of the watchful editorial 
 supervision which has been given to every scientific detail. Nothing 
 can well exceed the clearness and delicacy of the illustrative wood- 
 cuts. Altogether, the work may be said to have no parallel, either 
 in point of fulness or attraction, as a popular manual of physical 
 science. .... What we feel, however, bound to say, and 
 what we say with pleasure, is, that among works of its class no 
 publication can stand comparison either in literary completeness or 
 in artistic grace with it." Saturday Review. 
 
 THE APPLICATIONS OF PHYSICAL FORCES. By A. 
 GUILLEMIN. Translated from the French by Mrs. LOCKYER, and 
 Edited with Notes and Additions by J. N. LOCKYER, F.R.S. 
 Wi^h Coloured Plates and numerous Illustrations. Cheaper 
 Edition. Imperial 8vo. cloth, extra gilt. 2is. 
 Also in Eighteen Monthly Parts, price u. each. Part I. in October, 
 1878. 
 
 " A book which we can heartily recommend, both on account of the 
 width and soundness of its contents, and also because of the excel- 
 lence of its print, its illustrations, and external appearance." 
 "Westminster Review. 
 
io SCIENTIFIC CATALOGUE. 
 
 Hanbury. SCIENCE PAPERS : chiefly Pharmacological and 
 Botanical. By DANIEL HANBURY, F.R.S. Edited, with 
 Memoir, by J. INCE, F.L.S., and Portrait engraved by C. H. 
 JEENS. 8vo. i^s. 
 
 Henslow. THE THEORY OF EVOLUTION OF LIVING 
 THINGS, and Application of the Principles of Evolution to 
 Religion considered as Illustrative of the Wisdom and Benefi- 
 cence of the Almighty. By the Rev. GEORGE HENSLOW, 
 M.A., F.L.S. Crown 8vo. 6s. 
 
 " In one thing Mr. Henslow has done great good: he has shown 
 that it is consistent with a full dogmatic belief to hold opinions 
 very different from those taught as Natural Theology some half- 
 century ago" Nature. 
 
 Hooker. Works by Sir J. D. HOOKER, K.C.S.I., C.B., 
 M.D., D.C.L., President of the Royal Society : 
 
 THE STUDENT'S FLORA OF THE BRITISH ISLANDS. 
 Second Edition, revised and improved. Globe 8vo. los. bd. 
 
 The object of this work is to supply students and field-botanists with a 
 fuller account of the Plants of the British Islands than the manuals 
 hitherto in use aim at giving. " Certainly the fullest and most 
 accurate manual of the kind that has yet appeared. Dr. Hooker 
 has shown his characteristic industry and ability in the care and 
 skill which he has thrown into the characters of the plants. These 
 are to a great extent original, and are really admirable for their 
 combination of clearness, brevity, and completeness." Pall Mall 
 Gazette. 
 
 PRIMER OF BOTANY. With Illustrations. i8mo. is. New 
 Edition, revised and corrected. 
 
 Huxley and Martin. A COURSE OF PRACTICAL IN- 
 STRUCTION IN ELEMENTARY BIOLOGY. By T. H. 
 HUXLEY, LL.D., Sec. R.S., assisted by H. N. MARTIN, B.A., 
 M.B., D.Sc., Fellow of Christ's College, Cambridge. Crown 8vo. 
 6s. 
 " This is the most thoroughly valuable book to teachers and students 
 
 of biology which has ever appeared in the English tongue" 
 
 London Quarterly Review. 
 
 Huxley (Professor). LAY SERMONS, ADDRESSES, 
 AND REVIEWS. By T. H. HUXLEY, LL.D., F.R.S. New 
 and Cheaper Edition. Crown 8vo. Is. 6d. 
 
 Fourteen Discourses on the following subjects: (i) On the Advisable- 
 ness of Improving Natural Knowledge: (2) Emancipation 
 Black and White : (3) A Liberal Education, and where to find 
 #; (4) Scientific Education: (5) On the Educational Value of 
 
PHYSICAL SCIENCE. II 
 
 Huxley (Professor) continued. 
 
 the Natural History Sciences: (6) On the Study of Zoology: 
 (7) On the Physical Basis of Life: (8) The Scientific Aspects of 
 Positivism: (9) On a Piece of Chalk: (10) Geological Contem- 
 poraneity and Persistent Types of Life : ( 1 1 ) Geological Reform : 
 (12) The Origin of Species: (13) Criticisms on the " Origin of 
 Species:" (14) On Descartes' " Discourse touching the Method of 
 using One's Reason rightlv and of seeking Scientific Truth." 
 
 ESSAYS SELECTED FROM "LAY SERMONS, AD- 
 DRESSES, AND REVIEWS." Second Edition. Crown 8vo. u. 
 
 CRITIQUES AND ADDRESSES. 8vo. IQS. 6d. 
 
 Contents: I. Administrative Nihilism. 2. The School Boards: 
 what they can do, and what they may do. 3. On Medical Edu- 
 cation. 4. Yeast. 5. On the Formation of Coal. 6. On Coral 
 and Coral Reefs. 7. On the Methods and Results of Ethnology. 
 8. On some Fixed Points in British Ethnology. 9. Paleontology 
 and the Doctrine of Evolution. IO. Biogenesis and A biogenesis. 
 II. Mr. Darwin's Critics. 12. The Genealogy of Animals. 
 13. Bishop Berkeley on the Metaphysics of Sensation. 
 
 LESSONS IN ELEMENTARY PHYSIOLOGY. With numerous 
 Illustrations. New Edition. Fcap. 8vo. 4^. 6d. 
 This book describes and explains, in a series of graduated lessons, the 
 principles of Human Physiology, or the Structure and Functions 
 of the Human Body. ''Pure gold throughout" Guardian. 
 " Unquestionably the clearest and most complete elementary treatise 
 on this subject that we possess in any language." Westminster 
 Review. 
 AMERICAN ADDRESSES: with a Lecture on the Study of 
 
 Biology. 8vo. 6s. 6d. 
 
 PHYSIOGRAPHY: An Introduction to the Study of Nature. With 
 Coloured Plates and numerous Woodcuts. New Edition. Crown 
 8vo. is. 6d. 
 
 Jellet (John H., B.D.) A TREATISE ON THE 
 THEORY OF FRICTION. By JOHN H. JELLET, B.D., 
 Senior Fellow of Trinity College, Dublin ; President of the Royal 
 Irish Academy. 8vo. Ss. 6d. 
 
 Jones. THE OWENS COLLEGE JUNIOR COURSE OF 
 PRACTICAL CHEMISTRY. By FRANCIS JONES, Chemical 
 Master in the Grammar School, Manchester. With Preface by 
 Professor ROSCOE. New Edition. i8mo. with Illustrations. 2s. 6d. 
 
 Kingsley. GLAUCUS : OR, THE WONDERS OF THE 
 SHORE. By CHARLES KINGSLEY, Canon of Westminster. 
 New Edition, with numerous Coloured Plates. Crown 8vo. 6s. 
 
12 SCIENTIFIC CATALOGUE. 
 
 Langdon. THE APPLICATION OF ELECTRICITY TO 
 
 RAILWAY WORKING. By W. E. LANGDON, Member of the 
 Society of Telegraph Engineers. With numerous Illustrations. 
 Extra fcap. 8vo. 4^. bd. 
 
 Lockyer (J. N.) Works by J. NORMAN LOCKYER, F.R.S. 
 ELEMENTARY LESSONS IN ASTRONOMY. With nu- 
 merous Illustrations. New Edition. i8mo. $s. 6d. 
 " 77/i? book is full, clear, sound, and worthy of attention, not only as 
 a popular exposition, but as a scientific 'Index.'" Athenaeum. 
 " The most fascinating of elementary books on the Sciences" 
 Nonconformist. 
 
 THE SPECTROSCOPE AND ITS APPLICATIONS. By J. 
 NORMAN LOCKYER, F. R.S. With Coloured Plate and numerous 
 Illustrations. Second Edition. Crown 8vo. 3-r. 6d. 
 CONTRIBUTIONS TO SOLAR PHYSICS. By J. NORMAN 
 LOCKYER, F.R. S. I. A Popular Account of Inquiries into the 
 Physical Constitution of the Sun, with especial reference to Recent 
 Spectroscopic Researches. II. Communications to the Royal 
 Society of London and the French Academy of Sciences, with 
 Notes. Illustrated by 7 Coloured Lithographic Plates and 175 
 Woodcuts. Royal 8vo. cloth, extra gilt, price 3U. 6J. 
 " The first part of the work, presenting the reader with a continuous 
 sketch of the historv of the various inquiries into the physical con- 
 stitution of the sun, cannot fail to be of interest to all who care for 
 the revelations of modern science ; and the interest will be enhanced 
 by the excellence of the numerous illustrations by which it is accom- 
 panied" Athenaeum. " The book may be taken as an authentic 
 exposition of the present state of science in connection with the im- 
 portant subject of Spectroscopic analysis. . . . Even the unscientific 
 public may derive much information from it" Daily News. 
 PRIMER OF ASTRONOMY. With Illustrations. i8mo. is. 
 Lockyer and Seabroke. STAR-GAZING: PAST AND 
 
 PRESENT. An Introduction to Instrumental Astronomy. By 
 J. N. LOCKYER, F.R.S. Expanded from Shorthand Notes of a 
 Course of Royal Institution Lectures with the assistance of G. M. 
 SEABROKE, F.R. A. S. With numerous Illustrations. Royal 8vo. 2U.- 
 Lubbock. Works by SIR JOHN LuBBOCK,M.P.,F.R.S., D.C.L. : 
 THE ORIGIN AND METAMORPHOSES OF INSECTS. 
 With Numerous Illustrations. Second Edition. Crown 8vo. 3-r. 6d. 
 "As a summary of the phenomena of insect metamorphoses his little 
 book is of great value, and will be read with interest and profit 
 by all students of natural history. The whole chapter on the 
 origin of insects is most interesting and valuable. The illustra- 
 tions are numerous and good" Westminster Review. 
 ON BRITISH WILD FLOWERS CONSIDERED IN RELA- 
 TION TO INSECTS. With Numerous Illustrations. Second 
 . Edition. Crown 8vo. 4^. 6d. 
 
PHYSICAL SCIENCE. 13 
 
 Macmillan (Rev. Hugh). For other Works by the same 
 
 Author, see THEOLOGICAL CATALOGUE. 
 
 HOLIDAYS ON HIGH LANDS ; or, Rambles and Incidents in 
 search of Alpine Plants. Globe 8vo. cloth. 6s. 
 "One of the most charming books of its kind ever written" 
 Literary Churchman. "Mr. Macmillaris glowing pictures of 
 Scandinavian scenery" Saturday Review. 
 
 FIRST FORMS OF VEGETATION. Second Edition, corrected 
 and enlarged, with Coloured Frontispiece and numerous Illustra- 
 tions. Globe 8vo. 6s. 
 
 The first edition of this book was published under the name of 
 "Footnotes from the Page of Nature; or, First Forms of Vegeta- 
 tion." This edition contains upwards of IOO pages of neiv 
 matter and ele7>en new illustrations. " Probably the best popular 
 guide to the study of mosses, lichens, and fungi ever written. Its 
 practical value at a help to the student and collector cannot be 
 exaggerated. " Manchester Examiner. 
 
 Mansfield (C. B.) Works by the late C. B. MANSFIELD : 
 A THEORY OF SALTS. A Treatise on the Constitution of 
 Bipolar (two-membered) Chemical Compounds. Crown 8vo. 14*. 
 
 AERIAL NAVIGATION. The Problem, with Hints for its 
 Solution. Edited by R. B. MANSFIELD. With a Preface by J. 
 M. LUDLOW. With Illustrations. Crown 8vo. IGJ. 6d. 
 
 Mayer and Barnard. LIGHT.. A Series of Simple, Enter- 
 taining, and Useful Experiments in the Phenomena of Light, for 
 the use of Students of every age. By A. M. MAYER and C. 
 BARNARD. With Illustrations. Crown 8vo. 2s. 6d. 
 
 Miall. STUDIES IN COMPARATIVE ANATOMY. No. I, 
 The Skull of the Crocodile. A Manual for Students. By L. C. 
 MIALL, Professor of Biology in Yorkshire College. 8vo. 2s. 6d. 
 
 Miller. THE ROMANCE OF ASTRONOMY. By R. KALLEY 
 MILLER, M.A., Fellow and Assistant Tutor of St. Peter's Col- 
 lege, Cambridge. Second Edition, revised and enlarged. Crown 
 8vo. 4-r. 6d. 
 
 Mivart (St. George). Works by ST. GEORGE MIVART, F.R.S. 
 &c., Lecturer in Comparative Anatomy at St. Mary's Hospital: 
 ON THE GENESIS OF SPECIES. Crown 8vo. Second 
 Edition, to which notes have been added in reference and reply to 
 Darwin's "Descent of Man." With numerous Illustrations, pp. 
 xv. 296. 9-r. 
 
 "/;* no work in the English language has this great controversy 
 been treated at once with the same broad and vigorous grasp of 
 facts, and the same liberal a nd candid temper." Saturday Review. 
 
14 SCIENTIFIC CATALOGUE. 
 
 Mivart (St. George) continued. 
 
 THE COMMON FROG. With Numerous Illustrations. Crown 
 8vo. 3-r. 6d. (Nature Series.) 
 
 " It is an able monogram of the Frog, and something more. It 
 throws valuable crosslights over wide portions of animated nature. 
 Wo2ild that such works were more plentiful.' 1 '' Quarterly Journal 
 of Science. 
 
 Muir. PRACTICAL CHEMISTRY FOR MEDICAL STU- 
 DENTS. Specially arranged for the first M. B. Course. By 
 M. M. PATTISON MUIR, F.R.S.E. Fcap. 8vo. is. 6d. 
 
 Murphy. HABIT AND INTELLIGENCE, in Connection with 
 the Laws of Matter and Force : A Series of Scientific Essays. 
 By JOSEPH JOHN MURPHY. New Edition. [/ the Press. 
 
 Nature. A WEEKLY ILLUSTRATED JOURNAL OF 
 
 SCIENCE. Published every Thursday. Price 6d. Monthly 
 
 Parts, 2s. and 2s. 6d. ; Half-yearly Volumes, 15*. Cases for binding 
 
 Vols. is. 6d. 
 
 " This able and well-edited Journal, which posts up the science of 
 the day promptly, and promises to be of signal service to students 
 
 and savants Scarcely any expressions that we can employ 
 
 would exaggerate our sense of the moral and theological value of 
 the work" British Quarterly Review. 
 
 Newcomb. POPULAR ASTRONOMY. By SIMON NEW- 
 COMB, LL.D., Professor U.S. Naval Observatory. With 112 
 Engravings and Five Maps of the Stars. 8vo. iSs. 
 "As affording a thoroughly reliable foundation for more advanced 
 
 reading, Professor NevvcomVs ' Popular Astronomy ' is deserving 
 
 of strong recommendation" Nature. 
 
 Oliver Works by DANIEL OLIVER, F.R.S., F.L.S., Professor of 
 
 Botany in University College, London, and Keeper of the Herba- 
 rium and Library of the Royal Gardens, Kew : 
 LESSONS IN ELEMENTARY BOTANY. With nearly Two 
 Hundred Illustrations. New Edition. Fcap. 8vo. 4*. 6d. 
 This book is designed to teach the elements of Botany on Professor 
 Henslcntfs plan of selected Types and by the use of Schedules. The 
 earlier chapters, embracing the elements of Structural and Physio- 
 logical Botany, introduce us to the methodical study of the Ordinal 
 Tvpes. The concluding chapters are entitled, " How to Dry 
 Plants " and "How to Describe Plants." A valuable Glossary is 
 appended to the volume. In the preparation of this work free ust 
 has been made of the manuscript materials of the late Professor 
 Henslow. 
 
PHYSICAL SCIENCE. 
 
 Oliver continued. 
 
 FIRST BOOK OF INDIAN BOTANY. With numerous 
 Illustrations. Extra fcap. 8vo. 6s. 6d. 
 
 " It contains a well- digested summary of all essential knowledge 
 pertaining to Indian Botany, wrought out in accordance with the 
 best principles of scientific arrangement" Allen's Indian Mail. 
 
 Pennington. NOTES ON THE BARROWS AND BONE 
 CAVES OF DERBYSHIRE. With an account of a Descent 
 into Elden Hole. By ROOKE PENNINGTON, B.A., LL.B., 
 F.G.S. 8vo. 6s. 
 
 Penrose (F. C.) ON A METHOD OF PREDICTING BY 
 GRAPHICAL CONSTRUCTION, OCCULTATIONS OF 
 STARS BY THE MOON, AND SOLAR ECLIPSES FOR 
 ANY GIVEN PLACE. Together with more rigorous methods 
 for the Accurate Calculation of Longitude. By F. C. PENROSE, 
 F.R.A.S. With Charts, Tables, &c. 410. I2J. 
 
 Perry. AN ELEMENTARY TREATISE ON STEAM. By 
 JOHN PERRY, B.E., Professor of Engineering. Imperial College of 
 Engineering, Yedo. With numerous Woodcuts, Numerical Ex- 
 amples, and Exercises. i8mo. 45. 6d. 
 
 " Mr. Perry has in this compact little volume brought together an 
 immense amount of information, new told, regarding steam and 
 its application, not the least of its merits being that it is suited to 
 the capacities alike of the tyro in engineering science or the better 
 grade of artisan" Iron. 
 
 Pickering. ELEMENTS OF PHYSICAL MANIPULATION. 
 By E. C. PICKERING, Thayer Professor of Physics in the Massa- 
 chusetts Institute of Technology. Part I., medium 8vo. icw. 6</. 
 Part II., los. 6d. 
 
 " We shall look with interest for the appearance of the second volume, 
 and when finished ' Physical Manipulation'' will no doubt be 
 considered the best and most complete text-book on the subject of 
 which it treats." Nature. 
 
 Prestwich. THE PAST AND FUTURE OF GEOLOGY. 
 
 An Inaugural Lecture, by J. PRESTWICH, M.A., F.R.S., &c., 
 Professor of Geology, Oxford. 8vo. 2s. 
 
 Radcliffe. PROTEUS : OR UNITY IN NATURE. By. C. 
 B. RADCLIFFE, M.D., Author of "Vital Motion as a mode of 
 Physical Motion. Second Edition. Svo. js. 6d. 
 
 Rendu. THE THEORY OF THE GLACIERS OF SAVOY. 
 
 By M. LE CHANOINE RENDU. Translated by A. WELLS, Q.C., 
 late President of the Alpine Club. To which are added, the Original 
 
16 SCIENTIFIC CATALOGUE. 
 
 Re n du continued. 
 
 Memoir and Supplementary Articles by Professors TAIT and Rus- 
 KIN. Edited with Introductory remarks by GEORGE FORBES, B.A., 
 Professor of Natural Philosophy in the Andersonian University; 
 Glasgow. 8vo. 7-r. 6d. 
 
 ROSCOC. Works by HENRY E. ROSCOE, F.R.S., Professor of 
 Chemistry in Owens College, Manchester : 
 
 LESSONS IN ELEMENTARY CHEMISTRY, INORGANIC 
 AND ORGANIC. With numerous Illustrations and Chromo- 
 litho of the Solar Spectrum, and of the Alkalis and Alkaline 
 Earths. New Edition. Fcap. 8vo. 4-r. 6d. 
 
 CHEMICAL PROBLEMS, adapted to the above by Professor 
 THORPE. Fifth Edition, with Key. 2s. 
 
 * ' We unhesitatingly pronounce it the best of all our elementary 
 treatises on Chemistry." Medical Times. 
 
 SPECTRUM ANALYSIS. Six Lectures, with Appendices, En- 
 gravings, Maps, and Chromolithographs. Royal 8vo. 2U. 
 A Third Edition of these popular Lectures, containing all the most 
 recent discoveries and several additional illustrations. " The 
 lectures themselves furnish a most admirable elementary treatise 
 on the subject, whilst by the insertion in appendices to each lecture 
 of extracts from the most important published memoirs, the author 
 has rendered it equally valuable as a text-book for advanced 
 students." Westminster Review. 
 PRIMER OF CHEMISTRY. Illustrated. i8mo. is. 
 
 Roscoe and Schorlemmer. A TREATISE ON CHE- 
 MISTRY. By PROFESSORS ROSCOE and SCHORLEMMER. Vol. L, 
 The Non-metallic Elements. With numerous Illustrations and 
 Portrait of Dalton. Medium 8 vo. 2is. 
 
 " Regarded as a treatise on the Non-metallic Elements, there can be 
 no doubt that this volume is incomparably the most satisfactory one 
 of which we are in possession" Spectator. 
 
 [Vol. II. in the Press. 
 
 Rumford (Count). THE LIFE AND COMPLETE WORKS 
 OF BENJAMIN THOMPSON, COUNT RUMFORD. With 
 Notices of his Daughter. By GEORGE ELLIS. With Portrait. 
 Five Vols. 8vo. 4/. 14*. 6d. 
 
 Schorlemmer. A MANUAL OF THE CHEMISTRY OF 
 THECARBON COMPOUNDS OR ORGANIC CHEMISTRY. 
 By C. SCHORLEMMER, F.R.S., Lecturer in Organic Chemistry in 
 Owens College, Manchester. 8vo. 14^. 
 "// appears to us to be as complete a manual of the metamorphoses of 
 
 carbon as could be at present produced, and it must prove eminently 
 
 useful to the chemical student.'' Athenaeum. 
 
PHYSICAL SCIENCE. 17 
 
 Shann. AN ELEMENTARY TREATISE ON HEAT, IN 
 RELATION TO STEAM AND THE STEAM ENGINE. 
 By G. SHANN, M.A. With Illustrations. Crown 8vo. 45. 6d. 
 
 Smith. HISTORIA FILICUM : An Exposition of the Nature, 
 Number, and Organography of Ferns, and Review of the Prin- 
 ciples upon which Genera are founded, and the Systems of Classifi- 
 cation of the principal Authors, with a new General Arrangement, 
 c. By J. SMITH. A.L.S., ex-Curator of the Royal Botanic 
 Garden, Kew. With Thirty Lithographic Plates by W. H. FITCH, 
 F.L.S. Crown Svo. I2s. 6d. 
 
 "No one anxious to work tip a thorough knowledge of ferns can 
 afford to do "without it" Gardener's Chronicle. 
 
 South Kensington Science Lectures. Vol. I. Con- 
 taining Lectures by Captain ABNEY, F.RS., Professor STOKES, 
 Professor KENNEDY, F. J. BRAMWELL, F.R.S., Professor G. 
 FORBES, H. C. SORBY, F.R.S., J. T. BOTTOMLEY, F.R.S.E., 
 S. H. VINES, B.Sc., and Professor CAREY FOSTER. Crown Svo. 
 6s. [Vol. II. nearly ready. 
 
 SpOttlSWOOde. POLARIZATION OF LIGHT. By W. 
 
 SPOTTISWOODE, F.R.S. With numerous Illustrations. Second 
 
 Edition. Crown Svo. 3-r. 6d. (Nature Series.) 
 
 " The illustrations are exceedingly well adapted to assist in making 
 
 the text comprehensible '." Athenaeum. "A char, trustworthy 
 
 manual" Standard. 
 
 Stewart (B.) Works by BALFOUR STEWART, F.R.S., Professor 
 
 of Natural Philosophy in Owens College, Manchester : 
 LESSONS IN ELEMENTARY PHYSICS. With numerous 
 Illustrations and Chromolithos of the Spectra of the Sun, Stars, 
 and Nebulas. New Edition. Fcap. Svo. 4^. 6d. 
 The Educational Times calls this the beau-ideal of a scientific text- 
 book, clear, accurate, and thorough." 
 
 PRIMER OF PHYSICS. With Illustrations. New Edition, with 
 Questions. iSmo. is. 
 
 Stewart and Tait. THE UNSEEN UNIVERSE: or, 
 
 Physical Speculations on a Future State. By BALFOUR STEWART, 
 
 F.R.S.,and P. G. TAIT, M.A. Sixth Edition. Crown Svo. 6s. 
 
 " 7 he book is one which well deserves the attention ofthoughtjul ana 
 
 religious readers. . . . It is a perfectly sobtr inquiry, on scientific 
 
 grounds, into the possibilities of a future existence." Guardian. 
 
 Tait. LECTURES ON SOME RECENT ADVANCES IN 
 PHYSICAL SCIENCE. By P. G. TAIT, M.A., Professor of 
 Philosophy in the University of Edinburgh. Second edition, 
 revised and enlarged, with the Lecture on Force delivered before 
 the British Association. Crown Svo. <)s. 
 B 
 
1 8 SCIENTIFIC CA TALOGUE. 
 
 Tanner. FIRST PRINCIPLES OF AGRICULTURE. By 
 
 HENRY TANNER, F.C.S., Professor of Agricultural Science, 
 University College, Aberystwith, Examiner in the Principles of 
 Agriculture under the Government Department of Science. i8mo, 
 is. 
 
 Taylor. SOUND AND MUSIC : A Non-Mathematical Trea- 
 tise on the Physical Constitution of Musical Sounds and Harmony, 
 including the Chief Acoustical Discoveries of Professor Helm- 
 holtz. By SEDLEY TAYLOR, M.A., late Fellow of Trinity Col- 
 ledge, Cambridge. Large crown 8vo. &s. 6d. 
 " In no previous scientific treatise do we remember so exhaustive and 
 
 so richly illustrated a description of forms of vibration and of 
 
 'wave-motion in fluids" Musical Standard. 
 
 Thomson. Works by SIR WYVILLE THOMSON, K.C.B., F.R.S. 
 THE DEPTHS OF THE SEA : An Account of the General 
 Results of the Dredging Cruises of H.M. SS. "Porcupine" and 
 "Lightning" during the Summers of 1868-69 and 70, under the 
 scientific direction of Dr. Carpenter, F.R.S., J. Gwyn Jeffreys, 
 F.R.S., and Sir Wyville Thomson, F.R.S. With nearly 100 
 Illustrations and 8 coloured Maps and Plans. Second Edition. 
 Royal 8vo. cloth, gilt. -\\s. 6d. 
 
 The Athenjeum says : " The book is full of interesting matter, and 
 is wrilten by a master of the art of popular exposition. It is 
 excellently illustrated, both coloured maps and woodcuts possessing 
 high merit. Those who have already become interested in dredging 
 operations will of course make a point of reading this work ; those 
 who wish to be pleasamly introduced to the subject, and rightly 
 to appreciate the news "which arrives from time to time from the 
 ' Challenger i ,' should not fail to seek instruction from it." 
 
 THE VOYAGE OF THE " CHALLENGER." THE ATLAN- 
 TIC. A Preliminary account of the Exploring Voyages of H.M.S. 
 "Challenger," during the year 1873 and the early part of 1876. 
 With numerous Illustrations, Coloured Maps & Charts, & Portrait 
 of the Author, engraved'.by C, H. JEENS. 2 Vols. Medium Svo. 42^, 
 The Times says : " // is right that the public should have some 
 authoritative account of the general results of the expedition, and 
 that as many of the ascertained data as may be accepted with con- 
 fidence should speedily find their place in the general body of 
 scientific knowledge. No one can be more competent than the 
 accomplished scientific chief of the expedition to satisfy the public in 
 this respect. . . . The paper, printing, and especially the numerous 
 illustrations, are of the highest quality. . . . We have rarely, if 
 ever, seen more beautiful specimens of wood engraving than abound 
 in this work. . . . Sir Wyville Thomson's style is particularly 
 attractive; he is easy and graceful, but vigorous and exceedingly 
 
PHYSICAL SCIENCE. 19 
 
 Thomson con tinned. 
 
 happy in the choice of language, and throughout the work there are 
 touches which show that science has not banished sentiment from 
 his bosom." 
 
 Thudichum and Dupre. A TREATISE ON THE 
 ORIGIN, NATURE, AND VARIETIES OF WINE. 
 Being a Complete Manual of Viticultu re an d GEnology. By J. L. 
 W. THUDICHUM, M.D., and AUGUST DUPRE, Ph.D., Lecturer on 
 Chemistry at Westminster Hospital. Medium 8vo. cloth gilt. 25.5-. 
 
 tl A treatise almost unique for its usefulness either to the wine- grcnver, 
 the vendor, or the consumer of wine. The analyses of wine are 
 the most complete we have yet seen, exhibiting at a glance the 
 constituent principles of nearly all the wines known in this country." 
 Wine Trade Review. 
 
 Wallace (A. R.) Works by ALFRED RUSSET, WALLACE. 
 CONTRIBUTIONS TO THE THEORY OF NATURAL 
 SELECTION. A Series of Essays. New Edition, with 
 Corrections and Additions. Crown 8vo. Ss. 6d. 
 
 Dr. Hooker, in his address to the British Association, spoke thus 
 of the author : "Of Mr. Wallace and his many contributions 
 to philosophical biology it is not easy to speak without enthu- 
 siasm; for, putting aside their great merits, he, throughout his 
 writings, with a modesty as rare as I believe it to be uncon- 
 scious, forgets //is cnvn unquestioned claim to the honour of 
 having originated independently of Mr. Darivin, the theories 
 which he so ably defends." The Saturday Review says : " He 
 has combined an abundance of fresh and original facts with a 
 liveliness and sagacity of reasoning which are not often displayed 
 so effectively on so small a scale.'" 
 
 THE GEOGRAPHICAL DISTRIBUTION OF ANIMALS, 
 with a study of the Relations of Living and Extinct Faunas as 
 Elucidating the Past Changes of the Earth's Surface. 2 vols. 8vo. 
 with Maps, and numerous Illustrations by Zwecker, 42^. 
 The Times says: " Altogether it is a wonderful and fascinating 
 story, ^vhatever objections may be taken to theories founded upon 
 it. Mr. Wallace has not attempted to add to its interest by any 
 adornments of style ; he has given a simple and clear statement of 
 intrinsically interesting facts, and what he considers to be legiti- 
 mate inductions Jrom them. Nattiralists ought to be grateful to 
 him for having undertaken so toilsome a task. The work, indeed, 
 is a credit to all concerned the author, the publishers, the artist 
 unfortunately now no more of the attractive illustrations last 
 but by no means least, Mr. Stanford's map-designer" 
 
20 SCIENTIFIC CATALOGUE. 
 
 Wallace (A. R.) continued. 
 TROPICAL NATURE : with other Essays. Svo. 12s. 
 
 " Nowhere amid the many descriptions of the tropics that have been 
 given is to be found a summary of the past history and actual 
 phenomena of the tropics which gives that ivhich is distinctive of 
 the phases of nature in them more clearly, shortly, and impres- 
 sively" Saturday Review. 
 
 Warington. THE WEEK OF CREATION; OR, THE 
 
 COSMOGONY OF GENESIS CONSIDERED IN ITS 
 RELATION TO MODERN SCIENCE. By GEORGE WAR- 
 INGTON, Author of "The Historic Character of the Pentateuch 
 Vindicated." Crown Svo. 4^. 6d. 
 
 Wilson. RELIGIO CHEMICI. By the late GEORGE WILSON, 
 M.D., F.R.S.E., Regius Professor of Technology in the University 
 of Edinburgh. With a Vignette beautifully engraved after a 
 design by Sir NOEL PATON. Crown Svo. 8s. 6d. 
 "A more fascinating volume" the Spectator say s, " has seldom 
 fallen into our hands" 
 
 Wilson (Daniel.) CALIBAN : a Critique on Shakespeare's 
 "Tempest" and "Midsummer Night's Dream." By DANIEL 
 WILSON, LL.D., Professor of History and English Literature in 
 University College, Toronto. Svo. IQJ. 6d. 
 
 ' ' The whole volume is most rich in the eloquence of thought and 
 imagination as well as of words. It is a choice contribution at 
 once to science, theology, religion, and literature." British 
 Quarterly Review. 
 
 Wright. METALS AND THEIR CHIEF INDUSTRIAL 
 APPLICATIONS. By C. ALDER WRIGHT, D.Sc., &c., Lec- 
 turer on Chemistry in St. Mary's Hospital School. Extra fcap. 
 Svo. 3-r. 6d. 
 
 WurtZ. A HISTORY OF CHEMICAL THEORY, from the 
 Age of Lavoisier down to the present time. By AD. WURTZ. 
 Translated by HENRY WATTS, F.R.S. Crown Svo. 6s. 
 " The discourse, as a resume of chemical theory and research, unites 
 singular luminousness and grasp. A feiu judicious notes are added 
 by the translator." -Pall Mall Gazette. " The treatment of the 
 subject is admirable, and the translator has evidently done his duty 
 most efficiently" Westminster Review. 
 
MENTAL AND MORAL PHILOSOPHY, ETC. 21 
 
 WORKS ON MENTAL AND MORAL 
 PHILOSOPHY, AND ALLIED SUBJECTS. 
 
 Aristotle. AN INTRODUCTION TO ARISTOTLE'S 
 
 RHETORIC. With Analysis, Notes, and Appendices. By E. 
 M. COPE, Trinity College, Cambridge. 8vo. 14^. 
 ARISTOTLE ON FALLACIES; OR, THE SOPHISTICI 
 ELENCHI. With a Translation and Notes by EDWARD POSTE, 
 M.A., Fellow of Oriel College, Oxford. Bvo. Ss. 6d. 
 
 Birks. Works by the Rev. T. R. BIRKS, Professor of Moral Philo- 
 sophy, Cambridge : 
 
 FIRST PRINCIPLES OF MORAL SCIENCE ; or, a First 
 Course of Lectures delivered in the University of Cambridge. 
 Crown 8vo. 8s. 6d. 
 
 This work treats of three topics all preliminary to the direct exposi* 
 tion of Moral Philosophy. These are the Certainty and Dignity 
 of Moral Science, its Spiritual Geography, or relation to other 
 main subjects of human thought, and its Formative Principles, or 
 some elementary triiths on which its "whole development must 
 depend. 
 
 MODERN UTILITARIANISM; or, The Systems of Paley, 
 Bentham, and Mill, Examined and Compared. Crown 8vo. 6s. 6d t 
 MODERN PHYSICAL FATALISM, AND THE DOCTRINE 
 OF EVOLUTION ; including an Examination of Herbert Spen- 
 cer's First Principles. Crown 8vo. 6s. 
 
 Boole. AN INVESTIGATION OF THE LAWS OF 
 THOUGHT, ON WHICH ARE FOUNDED THE 
 MATHEMATICAL THEORIES OF LOGIC AND PRO- 
 BABILITIES. By GEORGE BOOLE, LL.D., Professor of 
 Mathematics in the Queen's University, Ireland, &c. 8vo. 14?* 
 
 Butler. LECTURES ON THE HISTORY OF ANCIENT 
 PHILOSOPHY. By W. ARCHER BUTLER, late Professor of 
 Moral Philosophy in the University of Dublin. Edited from the 
 Author's MSS., with Notes, by WILLIAM HEPWORTH THOMP- 
 SON, M.A., Master of Trinity College, and Regius Professor of 
 Greek in the University of Cambridge. New and Cheaper Edition, 
 revised by the Editor. 8vo. 12s. 
 
 Caird. A CRITICAL ACCOUNT OF THE PHILOSOPHY 
 
 OF KANT. With an Historical Introduction. By E. CAIRO, 
 M.A., Professor of Moral Philosophy in the University of Glasgow. 
 8vo. iJ. 
 
22 SCIENTIFIC CATALOGUE. 
 
 Calderwood. Works by the Rev. HENRY CALDERWOOD, M.A., 
 LL.D., Professor of Moral Philosophy in the University of Edin- 
 burgh : 
 
 PHILOSOPHY OF THE INFINITE : A Treatise on Man's 
 Knowledge of the Infinite Being, in answer to Sir W. Hamilton 
 and Dr. Mansel. Cheaper Edition. 8vo. 7^. 6d. 
 "A book of great ability .... written in a clear style, and may 
 
 be easily understood by even those who are not versed in such 
 
 discussions." British Quarterly Review. 
 
 A HANDBOOK OF MORAL PHILOSOPHY. New Edition. 
 Crown 8vo. 6s. 
 " It is, we feel convinced, the best handbook on the subject, intellectually 
 
 and morally, and does infinite credit to its author.^ Standard. 
 
 "A compact and useful work, going- over a great deal of ground 
 
 in a manner adapted to suggest and facilitate further study. . . . 
 
 His book will be an assistance to many students outside his own 
 
 University of Edinburgh. Guardian. 
 
 Fiske. OUTLINES OF COSMIC PHILOSOPHY, BASED 
 ON THE DOCTRINE OF EVOLUTION, WITH CRITI- 
 CISMS ON THE POSITIVE PHILOSOPHY. By JOHN 
 FISKE, M.A., LL.B., formerly Lecturer on Philosophy at 
 Harvard University. 2 vols. 8vo. 25^. 
 
 " The work constitutes a very effective encyclopedia of the evolution- 
 ary philosophy, and is well worth the study of all who wish to see 
 at once the entire scope and purport of the scientific dogmatism of 
 the day. 1 " Saturday Review. 
 
 Jardine. THE ELEMENTS OF THE PSYCHOLOGY OF 
 
 COGNITION. By ROBERT JARDINE, B.D., D.Sc., Principal of 
 the General Assembly's College. Calcutta, and Fellow of the Uni- 
 versity of Calcutta. Crown 8vo. 6s. 6d. 
 
 Jevons. Works by W. STANLEY JEVONS, M.A., Professor of 
 
 Logic in Owens College, Manchester. 
 
 THE SUBSTITUTION OF SIMILARS, the True Principle of 
 Reasoning. Derived from a Modification of Aristotle's Dictum. 
 Fcap. 8vOi 2s. 6d. 
 
 "Mr. Jevons' book is very clear and intelligible, and quite worth con- 
 sulting. " Guardian. 
 
 THE PRINCIPLES OF SCIENCE. A Treatise on Logic and 
 Scientific Method. New and Cheaper Edition, revised. Crown 
 8vo. I2J. 6d. 
 
 "No one in future can be said to have any true knowledge of what 
 has been done in the way of logical and scientific method in 
 England, without having carefully studied Professor Jevoni 
 book. " Spectator. 
 
MENTAL AND MORAL PHILOSOPHY, ETC. 23 
 
 MacCOll. THE GREEK SCEPTICS, from Pyrrho to Sextus. 
 An Essay which obtained the Hare Prize in the year 1868. By 
 NORMAN MACCOLL, B.A., Scholar of Downing College, Cam- 
 bridge. Crown Svo. 3.?. 6d. 
 
 M'Cosh Works by JAMES M'Cosn, LL.D., President of Princeton 
 
 College, New Jersey, U.S. 
 
 * ' He certainly shows himself skilful in that application of logic to 
 psychology, in that inductive science of the human mind which is 
 the fine side of English philosophy. His philosophy as a whole is 
 worthy of attention" Revue de Deux Mondes. 
 
 THE METHOD OF THE DIVINE GOVERNMENT, Physical 
 and Moral. Tenth Edition. Svo. los. 6d. 
 
 "This work is distinguished from other similar ones by its being 
 based upon a thorough study of physical science, and an accurate 
 knowledge of its present condition, and by its entering in a 
 deeper and more unfettered manner than its predecessors upon the dis- 
 cussion of the appropriate psychological, ethical, and theological ques* 
 tions. The author keeps aloof at once from the a priori idealism and 
 dreaminess of German speculation since Schelling, and from the 
 onesidedness and narrowness of the empiricism and positivism 
 which have so prevailed in England." Dr. Ulrici, in ' ' Zeitschrift 
 fur Philosophic." 
 
 THE INTUITIONS OF THE MIND. A New Edition. Svo. 
 cloth. ioj. 6d. 
 
 "The undertaking to adjust the claims of the sensational and in- 
 tuitional philosophies, and of the a posteriori and a. priori methods, 
 is accomplished in this work with a great amount of success." 
 Westminster Review. "/ value it for its large acquaintance 
 with English Philosophy, which has not led him to neglect the 
 %reat German works. I admire the moderation and clearness, as 
 well as comprehensiveness, of the author 1 s views " Dr. Dorner, of 
 Berlin. 
 
 AN EXAMINATION OF MR. J. S. MILL'S PHILOSOPHY: 
 Being a Defence of Fundamental Truth. Second edition, with 
 additions. los. 6d. 
 
 'Such a work greatly needed to be done, and the author was the man 
 to do it. This volume is ipiportant, not merely in reference to the 
 views of Mr. Mill, but of the whole school of writers, past and 
 present, British and Continental, he so ably represents" Princeton 
 Review. 
 
 THE LAWS OF DISCURSIVE THOUGHT : Being a Text- 
 book of Formal Logic. Crown Svo. 5_y. 
 
 " The amount of summarized information which it contains is very 
 great; and it is the only work on the very important subject with 
 which it deals. Never was such a work so much needed as in 
 the present day " London Quarterly Review. 
 
24 SCIENTIFIC CATALOGUE. 
 
 M ' C O Sh continued. 
 
 CHRISTIANITY AND POSITIVISM : A Series of Lectures to 
 the Times on Natural Theology and Apologetics. Crown 8vo. 
 7J. 6d. 
 
 THE SCOTTISH PHILOSOPHY FROM HUTCHESON TO 
 
 ' HAMILTON, Biographical, Critical, Expository. Royal 8vo. i6s. 
 
 Masson. RECENT BRITISH PHILOSOPHY .- A Review 
 
 with Criticisms ; including some Comments on Mr. Mill's Answer 
 to Sir William Hamilton. By DAVID MASSON, M.A., Professor 
 of Rhetoric and English Literature in the University of Edinburgh. 
 Third Edition, with an Additional Chapter. Crown 8vo. 6s. 
 " We can nowhere point to a work zvhich gives so clear an exposi- 
 tion of the course of philosophical speculation in Britain during 
 the past century, or which indicates so instructively the mutual in- 
 fluences of philosophic and scientific thought. " Fortnightly Review. 
 
 Maudsley. Works by H. MAUDSLEY, M.D., Professor of Medical 
 Jurisprudence in University College, London. 
 
 THE PHYSIOLOGY OF MIND ; being the First Part of a Third 
 Edition, Revised, Enlarged, and in great part Rewritten, of "The 
 Physiology and Pathology of Mind." Crown 8vo. los. 6d. 
 
 BODY AND MIND : an Inquiry into their Connexion and Mutual 
 Influence, specially with reference to Mental Disorders. An 
 Enlarged and Revised edition. To which are added, Psychological 
 Essays. Crown Svo. 6s. 6d* 
 
 Maurice. Works by the Rev. FREDERICK DENISON MAURICE, 
 M.A., Professor of Moral Philosophy in the University of Cam- 
 bridge. (For other Works by the same Author, see THEOLOGICAL 
 CATALOGUE.) 
 
 SOCIAL MORALITY. Twenty-one Lectures delivered in the 
 University of Cambridge. New and Cheaper Edition. Crown 8vo* 
 los. 6d. 
 
 " Whilst reading it ive are charmed by the freedom from exclusiveness 
 and prejudice, the large charity, the loftiness of thought, the eager- 
 ness to recognize and appreciate whatever there is of real worth 
 extant in the world, which animates it from one end to the other. 
 We gain new thotights and new ways of viewing things, even more, 
 pei'haps, from being brought for a time under the influence of so 
 noble and spiritual a mind" Athenaeum. 
 
 THE CONSCIENCE : Lectures on Casuistry, delivered in the Uni- 
 versity of Cambridge. New and Cheaper Edition. Crown Svo. 5-r. 
 The Saturday Review says : "We rise from them with detestation 
 of all that is selfish and mean, and with a living impression that 
 there is such a thing as goodness after all" 
 
MENTAL AND MORAL PHILOSOPHY, ETC. 25 
 
 Maurice continued. 
 
 MORAL AND METAPHYSICAL PHILOSOPHY. Vol. I. 
 Ancient Philosophy from the First to the Thirteenth Centuries ; 
 Vol. II. the Fourteenth Century and the French Revolution, with 
 a glimpse into the Nineteenth Century. New Edition and 
 Preface. 2 Vols. 8vo. 25^. 
 
 Morgan. ANCIENT SOCIETY : or Researches in the Lines of 
 Human Progress, from Savagery, through Barbarism to Civilisation. 
 By LEWIS ^H. MORGAN, Member of the National Academy of 
 Sciences. 8vo. i6s. 
 
 Murphy, THE SCIENTIFIC BASES OF FAITH. By 
 
 JOSEPH JOHN MURPHY, Author of " Habit and Intelligence." 
 
 8vo. I4J. 
 
 " The book is not without substantial value ; the writer contimies th e 
 work of the best apologists of the last century, it may be with less 
 force and clearness, but still ^vith commendable persuasiveness and 
 tact ; and with an intelligent feeling for the changed conditions of 
 the problem. " Academy. 
 
 Picton. THE MYSTERY OF MATTER AND OTHER 
 ESSAYS. By J. ALLANSON PICTON, Author of " New Theories 
 and the Old Faith. 3 ' Cheaper issue with New Preface. Crown 
 8vo. 6.r. 
 
 CONTENTS : The Mystery of Matter The Philosophy of Igno- 
 rance The Antithesis of Faith and Sight The Essential Nature 
 of Religion Christian Pantheism. 
 
 Sidgwick. THE METHODS OF ETHICS. By HENRY 
 
 SIDGWICK, M.A., Prselector in Moral and Political Philosophy in 
 Trinity College, Cambridge. . Second Edition, revised throughout 
 with important additions. 8vo. 14?. 
 A SUPPLEMENT to the First Edition, containing all the important 
 
 additions and alterations in the Second. 8vo. 2s. 
 " This excellent and very welcome volume Leaving to meta- 
 physicians any further discussion thaf may be needed respecting the 
 already over-discussed problem of the origin of the moral faculty, he 
 takes it for granted as readily as the geometrician takes space for 
 granted, or the physicist the existence of matter. But he takes little 
 else for granted, and defining ethics as ' the science of conduct, ' he 
 carefully examines, not the various ethical systems that have been 
 propounded by Aristotle and Aristotle s followers downwards, but 
 the principles upon which, so far as they confine themselves to the 
 strict provincf of ethics, they are based."' 1 Athenaeum, 
 c 
 
26 SCIENTIFIC CATALOGUE. 
 
 Thornton. OLD-FASHIONED ETHICS, AND COMMON- 
 SENSE METAPHYSICS, with some of their Applications. By 
 WILLIAM THOMAS THORNTON, Author of "A Treatise on Labour." 
 8vo. los. 6d. 
 
 The present volume deals with problems which are agitating the 
 minds of all thoughtful men. The following are the Contents : 
 /. Ante-Utilitarianism. II. History's Scientific Pretensions. III. 
 David Hume as a Metaphysician. IV. Huxleyism. V. Recent 
 Phase of Scientific Atheism, VI. Limits of Demonstrable Theism. 
 
 Thring (E., M.A.) THOUGHTS ON LIFE-SCIENCE. 
 By EDWARD THRING, M.A. (Benjamin Place), Head Master of 
 Uppingham School. New Edition, enlarged and revised. Crown 
 8vo. 7.5-. 6d. 
 
 Venn. THE LOGIC OF CHANCE : An Essay on the Founda- 
 tions and Province of the Theory of Probability, with especial 
 reference to its logical bearings, and its application to Moral and 
 Social Science. By JOHN VENN, M.A., Fellow and Lecturer of 
 Gonville and Caius College, Cambridge. Second Edition, re- 
 written and greatly enlarged. Crown Svo. ior. 6d. 
 " One of the most thoughtful and philosophical treatises on any sub- 
 ject connected with logic and evidence which has been produced in 
 this or any other country for many years.'" Mill's Logic, vol. ii. 
 p. 77. Seventh Edition. 
 
 LONDON I K. CLAY, SO^S, AND TAYLOR, PRINTERS. 
 
JLffWER D1YIMQN 
 
 
 South Kensington* 
 
 
 673230 
 
 V.I 
 
 
 LOWER DIVISION 
 
 UNIVERSITY OF CALIFORNIA LIBRARY