\\E UN!VER% 
 
 
 * 3 
 
 5 
 
 
 S 3 
 
 OF-CAU 
 
 \\E-UNIVER 
 
 clOS-ANCElfT 
 
 I S 
 
 % ,55*' 
 I i^ 1 
 
 ,OF-CAllf 

 
 5 6 
 .1 S 
 
 i I 
 
 x 

 
 DESCRIPTION 
 
 O F A 
 
 GLASS APPARATUS, 
 
 FOR MAKING 
 
 MINERAL WATERS, 
 
 LIKE THOSE opPYRMONT, SPA, SELTZER,&c, 
 In a few Minutes, and with a very little Expence : 
 
 TOGETHER WITH THE DESCRIPTION OF SOME 
 
 NEW EUDIOMETERS, 
 
 O R 
 
 INSTRUMENTS forafcertainingtheWnoLSOMENEss of 
 RESPIRABLE AIR; 
 
 AND THE 
 
 METHOD OF USING THESE INSTRUMENTS : 
 
 IN A LETTER TO THE 
 REV. DR. PRIESTLEY, LL.D. F.R.F; 
 BY J. H. DE MAGELLAN, F. R. S. 
 
 LONDON: 
 
 PRINTED FOR W. PARKER, No. 69, FLEET-STREET, 
 AND SOLD BY J. JOHNSON, No. ~2, ST. PAUL'S 
 CHURCH-YARD, AND W. BROWN, THE CORNER OF 
 
 3^SSiX-STRT, IN THE SI RAND. 
 MDCCLXXVII.
 
 Annex 
 
 5 
 oft 
 
 ADVERTISEMENT 
 
 OF THE 
 
 E I> I T O R, 
 
 RELATING TO THE USE OF THE SIM-- 
 PLE GLASS MACHINES FOR MAKING 
 MINERAL W A T E R S. 
 
 A LTHOUGH the method of ufing 
 jf~\. the fimple glafs machines of Mr. 
 Parker, is minutely defcribed in the fol- 
 lowing Letter, neverthelefs it may be 
 more agreeable to find here diftincT: refer- 
 ences to thofe articles, where it is parti- 
 cularly contained ; as the Author feem- 
 ingly could not help blending it with that- 
 of his improved machines.- 
 
 I. The defcription of the fimple Par- 
 ker's machine is comprifed No. 5 and 6. 
 
 II. The procefs to make ufe of the 
 fame, is defcribed No. 9, 10 and 1 1. 
 
 III. The manner of carrying on the 
 production of fixed air^ is indicated No,. 
 14 and note (</).. 
 
 IV.
 
 IV. The method for reducing the pro- 
 cefs to a few minutes is defcribed Ncr. 
 12, 13 ; fee note (c) and No. 15. 
 
 V. The manner of keeping thefe aci- 
 dulous waters is fhewn No. 16. 
 
 VI. For making them fparkle like 
 Champaign wine, No. 17. 
 
 VII. For making them ferruginous or 
 chalybeate, No. 18. 
 
 VIII. As to the medical and ceconomical 
 application of thefe waters, and fixed air^ 
 fee No. i, 3, 19 and 20. 
 
 N. B. Both glafs-machines-, for making arti- 
 ficial mineral waters, and the Eudiometers here de- 
 fcribed \ are made and fold WW. Parker's cut -giaft 
 i No. 69, Fleet fireet^ London. 
 
 ERRATA. 
 
 Add to the end of page 3, thefe words, Set Witbtrft 
 Qbftwathns on Chronic Weaknefi, page 164, Yoik, 1777, . 
 in Svo. 
 
 Page 39, laft line but one : two or tbne j read one or 
 tivo. 
 
 Page 45, about the middle : ||-| ; read |||,
 
 THE 
 
 CONTENTS. 
 
 ON ARTIFICIAL MINERAL WATERS. 
 
 Advertisement concerning the Jimple glafs ma- 
 chines for making thefe artificial waters Page iii 
 
 Their difcovery No. i 
 
 Different methods and injlruments for making 
 artificial mineral waters 2 
 
 The defcription of the Author's apparatus for 
 the fame 5 
 
 The Procefs 9 
 
 Powdered white marble better than chalk ; 
 See the note (b] to No. 9 
 
 Parker's machines hew may be ufed to greater 
 advantage-, Note (c} to No. 13 
 
 Qualities of thefe artificial waters 1 5
 
 . To make them fparkk &o. ij 
 
 To make them chalybeate 1 8 
 
 To ttfe them as a vehicle for medicines 19 
 
 To improve other Kquors byfxed air 2d 
 
 ON EUDIOMETERS. 
 
 Inattention of philofophers to fo important aft 
 .etyeft 2 
 
 Firjt attempts for conftrucling thefe inftru- 
 ments 2 z 
 
 Their ufefulnefs ; Note (e} to No. 22 
 
 Wby marjhy grounds are unfit for being inha- 
 bited ibid. 
 
 Defcription of thejrft new Eudiometer 24 
 
 Water preferable to mercury in thefe experi- 
 ments . 2 5 
 
 Tie Procefs 26 
 
 to fend atmofpherical air to be examined 
 elftwhere ; Note (f) to NQ. 26 
 
 Niceties
 
 ( vii ) 
 
 Niceties to be olftried in thefe Experiments ; 
 Note (g) to No. 26 
 
 A dejideratum to bring them to perfection; 
 ffote (b) to No. 27 
 
 Beft dimenjtons of a wooden trough for the 
 fame experiments ibid. 
 
 How to extratt the utmoft nitrous air 4 ibid. 
 
 Ufefulnefs of nitrous air for preferring from 
 Corruption ibid. 
 
 Phenomenon of the contraction and expanfion 
 cf the rf/r, independent from heat and cold\ 
 Note (?) to No. 28 
 
 How tq ejlimate the comparative falubrity of 
 air 3I 
 
 Attention required to the temperature of the 
 fltmofphere 35 
 
 Defcription of the fecond Eudiometer 3 6 
 
 The Procefs 37 
 
 Relative advantages of the three Eudiome- 
 tfrs 40 
 
 De-
 
 ( viii ) 
 
 Defcription of the third Eudiometer No. 41 
 The Procefs 43 
 
 Ufefulnefs of tbefe inftruments in many other 
 cafes 45 
 
 How to try dephlogijlicated air 46 
 
 Wonderful phenomenon of its diminution by 
 nitrous air 49 
 
 fhe circumftances relating to the fame 50 
 fynclufion 52 
 
 TO
 
 T O T H E 
 
 REVEREND DR. PRIESTLEY. 
 
 DEAR SIR, 
 
 I Do not know how better to employ the lei- 
 fure of thefe holydays I enjoy in your 
 neighbourhood, than in defcribing, according 
 to my promife (a), the two contrivances I have 
 mentioned in my laft letter ; which, I hope, 
 will be ufeful to the public. It is with plea- 
 fure I have obferved a great agreement in 
 almofl all our philofophical ideas : but I am 
 very happy to find chat we agree ftill more, in 
 
 (a) This refers to a former letter to* Dr. Prieftley, 
 printed in the Appendix to his third volume of Experi- 
 ments and Obfervationt on Air, No. Ill, p. 376 of the 
 London edition. 
 
 B look-
 
 looking with the greateft indifference on any 
 difcovery, even the moft ingenious, if no real 
 advantage may accrue from it to mankind. 
 Amongft the many that you have made, and 
 which are fcattered in your Philofophical Works, 
 that of producing by art,' at any time or place, 
 with very little expence and trouble, Mineral 
 Waters, like thofe of Pyrmont, Spa, Saltzer, 
 &c. whofe virtues depend on their being im- 
 pregnated with fixed air ; and that of finding 
 out a general ftandard or teft for afcertaining 
 the greater or lefs falubrity of refpirable air, 
 in any place whatfoever, are, undoubtedly, the 
 moft beneficial. The fuccefs with which the 
 firft of thefe two difcoveries is employed, 
 wherever it is known, and the very intereft- 
 ing obfervations relating to the fecond, made 
 alfnoft all over Italy by the Chevalier Landri- 
 ani, with his Eudiometer, clearly evince the 
 truth of my affertion. 
 
 2. As foon as your pamphlet, containing 
 the method of making Pyrmont Water, fell 
 into my hands, in the year 1772, I lent abroad 
 a great many copies of it, to different parts 
 of Europe, where I have a literary correipon- 
 dence ; this having long been my cuftom, 
 whenever any ufeful difcovery comes to my 
 knowledge. I made then fome change or im- 
 provement in your method,, which rendered 
 4 the
 
 ( 3 ) 
 
 the manner of conveying the fixed air to water 
 fomewhat eafier. This was added in a note to 
 the French tranflation of it, made foon after 
 at Paris, from a copy that I had fcnt to that 
 great promoter of Natural Philofophy, Mon- 
 fieur Trudaine de Montigny, mentioned page 
 268 of your fecond vol. on Different Kinds of 
 Air. Sometime after, Mr. Blunt invented a 
 machine, which rendered this operation ftill 
 more eafy. One of this kind, made by him- 
 felf at Mr. Nairne's, I fent to a very judicious 
 lover of philofophical experiments at Turin, 
 the Marquis de Rofignan ; now ambaflador 
 of the King of Sardinia to the court of Ber- 
 lin. 
 
 3. Another conftrudlion of a glafs machine 
 for the fame purpofe, was publilhed by Dr. 
 Nooth, in the 6,5th vol. of Philofophical Tranf- 
 actions. But chis being very imperfect, was 
 afterwards improved by Mr. Parker i and you 
 have given an account of it in the fecond vo- 
 lume of your work, above mentioned, page 
 298, and following. A very great number, 
 above a ihouland, of thefe machines have been 
 fcnt to different parts, even to the'Eaft Indies : 
 and it is known that many perlbns have 'been 
 greatly benefired by the ute of thefe artificial 
 acidulous waters. 
 
 B 2 4. I
 
 ( 4 ) 
 
 4. I found, however, not long ago, that 
 the manner of concluding the procefs, as de- 
 fcribed in the printed directions fent with thefe 
 machines, was very inconvenient, on account 
 of its flow operation j it requiring four, and 
 even fix cr icven hours to get the water fully 
 impregnated with fixed air. This I felt the 
 more in November laft, on my being at his 
 Serene Highnefr's the Duke of Arenberg, 
 whom I have a right to call my Mecasnas, on 
 account of the many favours I have received 
 from him. Knowing this generous Prince to 
 be endowed with the befl difpoiitions that any 
 of his rank ever had, for encouraging and 
 giving his protection to all improvements and 
 difcoveries beneficial to mankind, this confi- 
 deration prompted me to lend from London 
 one of thefe improved machines to BrufTels for 
 the ufe of his Hi^hnefs ; and, on my trying it, 
 on my arriv al there, I felt, for the firft time, 
 how difagreeable it was to wait fo long for the 
 defired effect ; which could be foon compleat- 
 ed, if the firft method already mentioned, No. 2, 
 was employed : for which reafon I always had 
 made ufe of this iaft, in preference to any 
 other , as it requires but few minutes to com- 
 pleat the operation. I then confidered what 
 could be done to avoid this. At laft I con- 
 trived the following apparatus, confifting of 
 fome additional pieces, by which means the 
 
 whole
 
 ( 5 ) 
 
 whole operation is fo fhortened, as to take but 
 few minutes : and, at the fame time, the quan- 
 tity of the artificial water is increafed to dou- 
 ble of that which is impregnated, at one pro- 
 cefs, in the fimple glafs-machine, improved by 
 Mr. Parker. 
 
 A DESCRIPTION of tils APPARATUS. 
 
 5. Let ABC (fig. i,) be one of the im- 
 proved machines of Mr. Parker, Handing up- 
 on a wooden dim d , in order to avoid any 
 water, if fpilled, from falling on the table. 
 The middle vefTd B has a neck, which is in- 
 ferted into the mouth of the vtiTcl A, being 
 nicely ground air-tight to it. This lower neck 
 of the middle veflel B has a (topple V of glafs, 
 conipofed of two parts, both having holes, 
 fufficient to let a good quantity of air pafs 
 through them : between thele two parts is left 
 a fmall fpace, containing a plano-convex lens, 
 which afts like a valve, in letting the air pafs 
 from below upwards, and hindering its return 
 into the veflel A. 
 
 6. The upper veflel C terminates below in 
 a tube, marked 2, i, (fig. i,) which being 
 crooked, hinders the immediate paflage of the 
 bubbles of fixed air into the upper vefTd C, 
 
 before
 
 ( 6 ) 
 
 before they reach the furface of the water in 
 the veffel B. The veflcl C is alfo ground air- 
 tight to the upper neck of the middle veflel 
 B ; and has a ftopple w, fitted to its upper 
 mouth, which is perforated through the mid- 
 dle. This upper veflel C contains juft.half 
 the water that can be contained in the under 
 one B ; and the end (i,) of its crooked tube 
 (2, i,) goes no lower than the middle of the 
 lame veflel B. 
 
 7. Befides thefe, I have added the two other 
 veflels, G and H (fig. 2,) perfectly like thofe 
 marked B and C (fig. i,). The veflel H is 
 furnifhed with a ftopper i, equally perforated 
 as the other w, and contains half as much as 
 the veflel G. Thefe veflels are let upon the 
 wooden (land F : and the lower neck of the 
 veflel G is not only furnifhed with a valve and 
 ftopple, as already defcribed in the veflel B, 
 but is fitted and ground air-tight to the neck 
 of the fame veflel A. Each of the three vrf- 
 fels, A, B, and G, have an opening, m, n y 
 and/, with ground floppies, which are only 
 open when occafion requires, as will be men- 
 tioned hereafter. 
 
 8. I contrived the wooden fland K, (fig. 3,) 
 fo that a thick piece of glafs #, like a fmall 
 tumbler, be cemented in the top, after it has 
 
 been
 
 ( 7 ) 
 
 been ground air-tight to the under necks of 
 the veflels B and G. The form of this ftand 
 is eafily conceived by the engraved plate, fig. 
 the ^d ; it being plane at the bottom, turns 
 up in a kind of convexity z z towards its edge, 
 and has a round moulding o o. Figure the 4th 
 reprefents a wide glafs funnel q, which may 
 enter into the upper mouth of the vefTel A. 
 Figure the 5th reprefents a fmall phial p, which 
 ferves to meafure the quantity of the vitriolic 
 acid to be made ufe of. Fig. the 6th reprefents 
 a little trough of tin R, to mealure the pounded 
 chalk or marble, that is to be employed in 
 every procefs : and fig. the jth reprefents a 
 particular kind of floppies, the ufe of which 
 will be explained hereafter. 
 
 THE PROCESS. 
 
 9. Let fome dry chalk, as it comes out of 
 the earth, that is to fay raw, and without be- 
 ing burned in the fire ; or rather white marble, 
 which is much better for the purpofe (), be 
 
 reduced 
 
 (1) White marble being firfl granulated, or pounded 
 like coarfe fand, is much better for the purpofe, than the 
 pounded chalk : becaufe the aftion of the diluted acid 
 upon the marble lads a very confiderable time ; and the 
 the fupply of ibefxed air, which is difengaged by this 
 
 effer-
 
 8 
 
 reduced to powder ; and let fome oil of vitriol 
 be at hand The veffrl 5, together with C, 
 (fig. i,) muft be taken off from A, and put 
 into the (land K (fig. 3,). The vefiels C and 
 H being feparated from their under ones B and 
 G ; let thefe laft be filled with fpring, or any 
 other drinking water, or even with, diftilled 
 water ; and let them be joined again with the 
 upper veffels C and H. 
 
 10. Let fome water be poured on the lower 
 vefTel A, fo as to cover the rifing part of its 
 bottom : but if this appears too vague a di- 
 rection, pour in fourteen or fixteen meafures of 
 water, with the glafs p (fig. 5,) : then fill the 
 fame phial p with oil of vitriol, and pour it into 
 the fame veifel A, along with the water. 
 
 Now let the glafs funnel q (fig. 4,) be put 
 into the mouth of the veflel A ; and filling the 
 
 effervefcence, is much more regular than otherwife. In 
 general it continues to furtiifh faed air more than twenty- 
 four hours. When no more air is produced, if I decant out 
 of the veflel A, all the acid fluid, already faturated, and 
 wafli off the thin, white fediment, I may employ again the 
 remaining granulated marble, by adding to it frefh water, 
 and a new quantity of vitriolic acid ; which will then 
 furnifti a further fupply of fixed air : and this may be re- 
 peated over again, until all the marble is diflblved ; which 
 will not be very foon. 
 
 trough
 
 C 9 ) 
 
 trough R (fig. 6,) with the pounded chalk or 
 marble, let it be thrown into it. Take offthe 
 funnel , which is ufed only to prevent the 
 chalk from touching the infide of the mouth of 
 this vdTel : fince otherwife it will flick fo 
 ftrongly to the neck of the vefTel B, as not to 
 allow the taking it off again without breaking. 
 Then immediately place the two veflcls B and 
 C as they are, over the mouth of the veffel A 
 and all the fixed air which is difengaged from 
 the chalk or marble, by the force of the diluted 
 acidj will pafs up through the valve V into the 
 vefiel B. When this fixed air comes to the 
 top of the veiTel B, it will difiodge from thence 
 as much water as its bulk : and this water, fo 
 diflodged, will go up, by the crooked tube 2, i, 
 into the veflel C. 
 
 1 1. Care mud be taken not to {hake the vefiel 
 A when the powdered chalk is poured in , for 
 orherwife a great and fudden effervefcence 
 willenfue, which will, perhaps, expel part of 
 the contents. In fuch a cafe, it will be necef- 
 fary to open the ftoppSe ;;;, in order to give 
 vent to the efFerveicence tor a moment , other- 
 wife the vefTel A may happen to burft. Per- 
 haps it will be neceiTary to throw away the 
 contents, to wafh the veflcl with water (be- 
 ranfe the boiling matter will ftick ber.ween the 
 necks of thde vdfcls, and will cement them to- 
 C gether)
 
 gether) and to begin the operation afrefk But 
 if the powdered chalk is thrown in, without 
 any confiderable {hake of the machine, there 
 will be but a fmall effervefcence at the begin- 
 ning. When this operates well, the vefTel C 
 will foon be filled with water, and the vefiel 
 J5 half filled with air ; which when done will 
 be eafily perceived, by the air going up in 
 large bubbles by the crooked tube 1,2; this 
 will take place in about two or three minutes. 
 
 12. As foon as this is obferved, take off the 
 two vefiels B and C, together as they are, from 
 A-, put them on the (land (fig. 3,) and, taking 
 the other two, G H, from the {land F, put 
 them over the veflel A. Whilft the operation 
 is going on in the vefiels G H, hold, with the 
 right hand, the veflels B C, which are now on 
 the {land A', by the neck z z, inclining them a 
 little fidewaya, and make them very brifkly, 
 io that the water within B be very much agi- 
 tated, prefenting many frem furfaces in con- 
 tact with the fixed air, great part of which 
 will be abforbed into the water ; as it will ap- 
 pear by the end of the crooked tube being con- 
 fiderably under the furface of the water in the 
 vefiel B. 
 
 13. It will fuffice to fhake the water in this 
 manner during two or three minutes ; which 
 
 done,
 
 ( 'I ) 
 
 done, loofen the upper veflel C, fo that the re- 
 maining water may fall into the veflel B, and 
 the unabforbed air may go out.: then taking 
 off thefe veflels from the (land K, put them, 
 joined together as they are, on the (land F (c). 
 By this time the veflel G will be half filled 
 with fixed air, and the upper veflel H will be 
 filled with the water, thrown up by it. Take 
 then thefe yefiels to the ftand K, and replace 
 the others B C, on the mouth of the veflel A ; 
 in order to be half filled again with fixed air, 
 whilft the water in the veflels G H is brifkly 
 {baked in the fame manner as the others have 
 been. 
 
 14. Whenever the effervefcence nearly ceafes 
 in the veflel A, it will be revived again by 
 giving it a gentle Ihake, fo that fome part of 
 the powdered chalk which is in a heap at the 
 bottom of A, may be mixed with the diluted 
 vitriolic acid, and difengage more fixed air. 
 However, when it happens that the whole is 
 exhaufted, and no more air rulhes up to ths 
 middle veflel from the lower one, either more 
 powdered chalk muft be put in, or more oil 
 pf vitriol 5 or at, laft more water, if neither of 
 
 (0 By this method, even the fimple glafs machines 
 ?bove mentioned No. 3, already defcribe.d No. 5 an,4 
 6, may be worked fo as to have the water fully impreg- 
 pated in a ftw minutes, though with lefs advantage. 
 
 C 2 ' the
 
 the two firft produced the defired effect. Thefe, 
 additions may be performed by letting them 
 in, through the mouth of the vefTelA, whilft the 
 upper veflels are changing. In this cafe ufe 
 muft always be made of the funnel ^, in order 
 to avoid the fticking of the junctures above 
 mentioned (<). 
 
 15. When this operation has been repeated 
 three or at moft four times alternately, with each 
 iet of veffels, by throwing out the remaining air 
 which does not incorporate with the water, 
 and putting in a frefli quantity of fixed air, 
 the water contained in the two veflels B and 
 G, will be fully faturated ; and will be much 
 more pleafant to tafte than the natural Pyr- 
 mont or Seltzer's waters, which, befides their 
 fixed air (hardly the half of what this artificial 
 water may abforb) contain fome difagreeable 
 faline tafte, which, it is known, does not con- 
 tribute at all to their medicinal virtues ; but, 
 on the contrary, it may be hurtful in fome 
 complicated cafes. 
 
 (^) Thefe directions mult be refpeftively applied to the 
 fimple glafs machines, mentioned No. 3, and defcribed 
 No, j and 6, when they are employed. If thefe vefTeis be 
 buffered to ttandfix or feven hours, the water will be fuf~ 
 fictently impregnated without agitation j provided the fup~ 
 piy of fixed air be copious. 
 
 1 6. Thefe
 
 ( 13 ) 
 
 1 6. Thefe artificial waters will remain as 
 limpid and as tranfparent as before, although 
 there has been abforbed above as much air as 
 their own bulk. The whole procefs will hard- 
 Jy take aboye a quarter of an hour, by .this 
 method -, and the quantity will be double of 
 that which could be made in the fimple glafs- 
 machine. The water may be taken out by the 
 opening / or n, to be drank immediately ; if 
 not, it will be better to let it remain in the 
 machine, where it has no communication with 
 the external air : otherwife, the fixed air goes 
 off by degrees, and it becomes vapid and flat ; 
 as it happens alfo to the natural acidulous wa- 
 ters. This artificial water may be kept a long 
 time, in bottles well flopped, placed with their 
 mouths downwards. 
 
 17. In general they are fo fimilar to the na- 
 tural acidulous waters, that they may be even 
 made to fparkle like Champaign wine. Mr. 
 Warltire has actually brought thefe waters to 
 this flate, by keeping the fixed air compreflfcd 
 upon the furface of the water in the middle 
 veiTel ; as appears by his letter printed in the 
 Appendix to your third volume of Experiments 
 and Obfervations on Alr^ page 366. The fame 
 end will be obtained, if, inftead of the flop- 
 pies w and *', ufe is made of the folid one re- 
 prefemed (fig. ;,) which has a kind of bafon 
 
 at
 
 ( H ) 
 
 at the top, in order to hold fome additional 
 weight. This ftopple muft be of a conical 
 figure, and very loofe ; but fo well ground 
 and fmooth in its contact, as to be air-tight 
 by its preffure, which may be increafed by 
 fome additional weights in its bafon. If the 
 veflels are flout enough, there is, no danger 
 of their burfting in the operation, 
 
 1 8. Thefe waters may alfo be rendered fer- 
 ruginous (or chalybeate) very eafily, by put- 
 ting, in the middle vefiel, two or three balls of 
 fine iron-binding wire : otherwife two or three 
 ounces of fmall iron nails may be put in for 
 the fame purpofe ; becaufe they will diflblve 
 iron as much as to be faturated with it in a 
 few hours, according to the experiment of 
 Mr. Lane. According to Sir John Prinojle, 
 there may be added from eight to ten drops of 
 tinftiira martis cum fpiritu fatis, in order to 
 refemble more nearly the genuine Pyrmont: 
 water. 
 
 19. No doubt thefe waters may be advan- 
 tageoufly employed in many medical purpofes ; 
 not only by diflblving in them the very falts 
 which are found to be contained in many na- 
 tural fprings, renowned for their different vir- 
 tues, but even they may be applied as a ve- 
 hicle to many draughts and internal medicines, 
 
 whicr}
 
 ( '5 ) 
 
 which will be lefs naufeous to the patients, 
 and perhaps more agreeable to the ftomach, 
 giving to it a tonical ftrength. 
 
 20. I (hall conclude this fubject by obferv- 
 ing with you, that fixed air may be given to 
 wine, beer, cyder, and to almoft any liquor 
 \vhatfoever. Even when beer is become flat, 
 or dead as it is called, it may be revived by 
 employing the fame method : but the delicate, 
 though brifk, and agreeable flavour, or acidu- 
 lous tafte, communicated by the fixed air, and 
 which is fo manifeft in water, will hardly be 
 perceived in wine or other liquors which have 
 much tafte of their own. I fhall now proceed 
 to the fecond object of this letter. 
 
 ON EUDIOMETERS. 
 
 21. The happy difcovery you have made 
 for the general benefit to mankind, and per- 
 haps of almoft the whole animal creation of 
 this globe, by finding that nitrous air is a true 
 tejl of the purity of refpirable air, which is ab- 
 folutely neceffary to life, and without which 
 it is prefently extinct, gives a moft ftriking 
 inftance of the blameable flownefs of mankind 
 to pay a proper attention to thofe objects, the 
 importance of which is infinitely fuperior to 
 i that
 
 ( 16 ) 
 
 that of the numerous trifling novelties which 
 'fo often fpread with prodigious rapidity through" 
 remote provinces, and even to the moft diftant 
 countries of the earth. Since the beginning 
 of the year 1772, in which you announced 
 this moft interefting and valuable difcovery, 
 in the 6zd vol. of the Pbilofopbical Tranfafii- 
 cns, no more than three or four philofophers 
 that I know of, have given any confiderablc 
 degree of attention to fo important a fub- 
 jeft. 
 
 22. The Abbe Fontana and the Chevalier 
 Landriani, both of Italy, and already known 
 to the public by their feveral valuable pro- 
 ductions, were the firft, as it feems, who 
 availed themfelves of this difcovery (i). Both 
 
 propofed 
 
 (e) Mr. Volta, Profeflbr of Natural Philofophy at Co- 
 mo in Italy, has made a difcovery, mentioned No. V. of 
 the Appendix to your third volume On different Kinds of 
 Air, which feems clofely conne&ed with the prefent fub- 
 jel. He discovered that inflammable air is contained in 
 the mud of almoft all lakes, marfhes, and wet grounds cf 
 Italy. He publimed different letters on this fubjeft, of 
 which he was fo kind as to fend me a printed copy, after 
 part of this letter was printed. The experiments you have 
 made afterwards with ma at Calne on this matter, mow that 
 this air is lefs inflammable than what proceeds from the 
 folution of metals, with vitriolic acid : it burns with a 
 lambent flame, like the air produced by heat from char- 
 Ooai. This difcovery of profeffor Volta accounts very 
 
 well
 
 ( '7 ) 
 
 propofed to the public one of the moft ufeful 
 kinds of inftruments that we can boaft of 
 among the numberlefs ones already employed 
 in philofophical refcarches and experiments. 
 They gave to thefe inftruments different forms, 
 as appears by the printed defcriptions that each 
 of them has feparately publiihed : and the Che- 
 valier Landrianihas tranfmitted to England, as 
 a prefent to you, the very inftrument he had 
 made ufe of, to eftimate the reipective falu- 
 brity of the air in different parts of Italy, as 
 mentioned page 23 of the Preface to your 
 third volume On different Kinds cf Air. This 
 
 well for the unhealthy habitations fuch marfhy grounds 
 generally afford to the human fpecies : and fliows the ne- 
 ceflity of examining, with care, by means of the Eudiome- 
 ter, what places are fit for being inhabited. This is a 
 new and a very interefting requisite, never to be over- 
 looked, before any building is erefted, or the place for 
 any country feat is fixed upon. Such grounds or places 
 whofe atmofphere is loaded with phlogiftic miafma, are 
 the moft dangerous to animal life : becaufe the air of fucK 
 an atmofphere cannot be a good conductor or dilcharger of 
 the fuperabundant phlogifton, of which the animal ccco- 
 riomy requires to be unloaded : this being the aim intend- 
 ed by Nature in the function of refpiration, as you have 
 at laft difcovered, and incontrovertibly demonftrated, by 
 the moft decifive experiments, to be the cafe ; after fo ma- 
 ny ineffe&ual attempts of the greateft philofophers of all 
 ages. This appears by fedion V, page 55 and following, 
 of your third volume On different Kinds of Air. London 
 di;. 1777, and by Phil. Tran/aff, vol. 66, p. 226. 
 
 P E.U-
 
 Eudiometer is fmaller than that defcribed by 
 him, and published in the 6th volume of Ro- 
 fier's Journal for the year 1775, though near- 
 ly of the fame form. It confifts of a glafs 
 tube, ground to a cylindrical vefiTel, with two 
 glafs cock?, and a fmafl bafon, all fitted in a 
 wooden frame. Quickfilver is there ufed in- 
 ftead of water ; and that part of it which re- 
 places the bulk loft by the diminution of the 
 two mixed airs, is conducted either through 
 a kind of glafs fiphon, or through the capil- 
 lary holes of a giafs funnel : fo that by its 
 fall, the whole mixture of the two kinds of 
 air is more readily made. 
 
 23. Dr. Falconer pf Bath fent, fome time 
 ago, to the Royal Society of London, a glafs 
 tube, neatly divided ; by means of which one 
 may be enabled to know the quantity of dimi- 
 nution produced in a certain bulk of the mix r 
 ture of nitrous air with another air, in order to 
 judge of its falubrity, which you have (hewn 
 to be in proportion to the diini nution Buffered in 
 the fum of their original bulk^ after they are mix- 
 ed together. This method is the neareft to your 
 original one, or, rather, is the very fame you 
 have ufed in the purfuit of this difcovery ; as 
 appears by your printed work on this fubject : 
 and I think it to be the readieit of all, when- 
 ever no great nicety is required in obfervations 
 3 of
 
 ( '9 ) 
 
 bf this kind. There are^ however, fo many 
 circumftances necefTary in a good inftrument 
 for fully anfwering fo great ah object to its ut- 
 moft extent; that I ihould be deterred from 
 offering to the public what I have as yet done 
 on this i abject, was I not aware, that fome 
 advantages always accrue to public good, by 
 any new fteps towards perfection, how diflant 
 foever we may ftill happen to be from its com- 
 pieat attainment. 
 
 DESCRIPTION OF THE FIRST NEW 
 EUDIOMETER. 
 
 24. Of the three Eudiometers I have con- 
 trived, which are repreiented fig. 8, 15 and 
 1 6 in the annexed plate, I think thelaft is the 
 eafieft in its application, and the moft exact in 
 its refults. It is reprefented alfo (fig. 12, 14, and 
 1 7) in different pofitionSj for the better under- 
 ftanding of its application : and it confifts of 
 the following parrs : viz. a glafs tube m n e d 
 fig. 1 6, about twelve or fifteen inches long, 
 and of an equal diameter, with a ground glafs 
 ftopple, m : a vefTel, f, the neck of which is 
 ground air-tight to the lower end d of the 
 tube : and two equal phials a and , whofe 
 necks are alfo ground air-tight to the refpec- 
 tive mouths of the vefiel c. Both thefe phials 
 D 2 contain
 
 contain nearly as much as the whole tube rrt 
 n e d. There is, moreover, a fliding brafs ring, 
 marked z, which flides in the tube n d^ and 
 may be made tight at pleafure by a finger- 
 fcrew : and, laftly, a ruler, either of brafs or 
 of wood, reprefented fig. n, which is divided 
 into equal parts, and indicates the contents of 
 both the phials a and , when thrown into the 
 tube, by the number of parts which is en- 
 graved or ftamped about the middle of it. 
 The two bent pieces of brafs z t ferve to hold 
 k eafily by the fide of the tube n d fig. 14 and 
 1 7, keeping it dole to its neck by the notch ;'. 
 
 25. Experiments with thefe Eudiometers, 
 which are eafily conftructed, may be made 
 either with water or with quickfilver ; with 
 this difference, that when the laft is made ufe 
 of, the Eudiometers (particularly the third, 
 reprefented fig: 8, which feems the fitcft for 
 being ufed with quickfilver) will be more con- 
 venient if made of a ftill fmaller fize. Mer- 
 cury, however, is a fluid that, I think, never 
 ought to be ufed preferably to water, in the 
 infidc of Eudiometers ; becaufe it luffers a fen- 
 fible action from the contact with nitrous air, 
 as yourfelf have obferved : and this muft have 
 an influence on the refulc of the experiments. 
 Water, on the contrary, feems lefs liable to mif- 
 takes, although it imbibes fome part of the ni- 
 trous
 
 trom air. In fa<5l this effect only takes place 
 in a long time, or with much agitation : and, 
 after duly weighing the queilion on both 
 fides, I mould think water may be generally 
 ufrd, without the fear of any fenfible error. 
 The weight and the dearnefs of quick filver, 
 are, likewife, two other confiderations to give 
 the preference to water in thefe experiments. 
 
 THE PROCESS. 
 
 26. In the firft place there muft be either a 
 trough, as repreferited fig* 17 ; or at lead a 
 common tub, nearly filled up with water, un- 
 lefs the tall glafs receiver, of which I mail 
 fpeak No. 34, b? at hand. I take out the 
 Hopples (fig. 1 6,) fill the Eudiometer entirely 
 with water, keeping it in the pofition repre- 
 fentcd fig. 1 6 and 17. I then fhut it with the 
 ftopple m^ without leaving any bubble of air 
 in the infide ; and put the lower part c under 
 thefurface of the water in the tub (fig. 17) in 
 an erect pofition as it is therein feen. I take 
 the phial a, filled with water ; and, keeping 
 its mouth downwards under the furface of the 
 water, I fill it with that air, the falubrity of 
 which I want to afcertain (f). This is done 
 
 either 
 
 (f) The cafe I am fpeaking of, is when I have a bottle 
 
 of
 
 either by putting the phial a on the fhelf n s 
 of the tub fig. 17, and throwing the air into the 
 glals funnel /, which is there cemented to the 
 fhelf j or by holding in the left hand the fame 
 phial a together with the glafs funnel B 
 (which is reprefented fig. 18, and has no pipe 
 at all) applied to the mouth of the phial, 
 whilft I pour the air with my right hand into 
 it. But left the heat of my hand (hould pro- 
 duce any confiderable expanfion in this air, I 
 generally ufe, in hot weather, the wooden tongsj 
 reprefented fig. 21, with two bent wires x #, 
 
 of air, which has been taken at any diftarit place, and feht , 
 for trial. If a glafs bottle, with a ground glafs ftopplc, 
 is filled with water or with mercury, and emptied in the 
 place whofe atmofpherical air is intended for being examin- 
 ed, it will, of courfe, be filled with that air : and, beirig 
 clofely (hut with the glafs ftopple, may be carried to arty 
 diftant place for trial. By this means the atmofpherical 
 air of any part of a country may be fent to any diftant one, 
 in order to afcertain its comparative falubrity : and many 
 ufeful inquiries and difcoveries may be made hereafter 
 on this fubjeft, with great eafe, and at very flnall eX- 
 pence. 
 
 But if I only want to try the air of the room, where t 
 have the Eudiometer, I then only pour out of the phial a 
 the water it contains. I find that, however, after fome trials 
 with nitrous air, the atmofphere about me is loaded with 
 phlogiitic miafma : and for that reafon I always empty the 
 phial a out of the window of the room, in order to have 
 nearly the fame kind of air in all the experiments. 
 
 ia
 
 in order to hold the glafs funnel z clofe to the 
 mouth of the phials , unlefs they are made 
 vyith a folid lump at their bottoms, as repre- 
 fented in the plate. See note (^). 
 
 27. The phial a being filled with that air, 
 the falubricy of which I ^m to examine, I put it 
 into the mouth of the veflel c, making it rather 
 
 tight, 
 
 () There are fome niceties tq be obferved in order to fill 
 up, exaclly, any phial intended to fcrve as a meafure or* 
 air, of which I muft give an account in this place. The 
 eafiefl method to fucceed is the following : Let a glafs 
 funnel / (fig. 17) be cemented under the hole of the fhelf 
 n o in the trough. In this cafe I hold the phial a filled 
 vyith water, with its mouth downwards over the hole n of 
 the funnel /: I throw the air into the funnel ; and, when 
 the phial is filled with air, I take it fideways, rubbing its 
 mouth along the furface of the fhelf, fo that the redundant 
 air, adhering to the mouth of the phial, be got off: and 
 I put it into the mouth of the Eudiometer belonging to it. 
 But as the heat of the hand muft expand the air contained 
 in the phial, which of courfe will then contain lefs air 
 than its real meafure in the temperature of the furround- 
 ing water, I handle the phial with a kind of pliers or 
 tongs of wood, reprefented fig. 21, till the neck enters 
 into the proper place of the vefTel c, where 1 fecure it with 
 the other hand : and, laying afide the wooden tongs, I 
 paake it properly tight. But if the phials have a folid 
 knob at their bottoms, as reprefented in the plate, it will 
 then be enough to handle them by it only : fince the heat 
 of the hand can not be communicated in fo fhort time to 
 the air in the infide. 
 
 If
 
 tight, which muft be done with fome care ; 
 for if the phials a and b are not tight enough 
 to the refpective mouths of the veflel c, 
 they will flip out, when turned downwards, 
 and of courfe will be broken : and, if they 
 ?re too tight, the veflel c will beeafily cracked, 
 and become unfit for ufe, The better to 
 avoid thefe accidents, and to judge of the 
 proper degree of tightnefs, let the necks of 
 the phials a b, and of the vdfel f, as well as 
 the glafs ilopple m{ be alwa s rubbed with 
 tallow, previoufly to every experiment. When 
 I have done with the phial <*, I take the other 
 phial , filled with water : by the fame method 
 
 If I have not the convenience of a trough, prepared 
 with a fhelf, and its fixed funnel, as above roehliqned, 
 an affiftant holds the funnel under the water in a common 
 tub, whilft I fill up the phial with air: and I take care to 
 bold the phiai in fuch a manner that the end of the funnel 
 be out of the infide of the phial at the lafl moment, that 
 the air may rum out after it is totally filled : otherwife 
 that part of the phial, occupied by the end of the funnel, 
 will not be totally filled with the air. 
 
 Even without any afliftant, but with a little care, a per- 
 fon may hold both the phial and the funnel in the left 
 band, whilft he throws the air into it with the other hand ; 
 as I have myfelf frequently done in experiments of this 
 kind : and when I make ufe of the wooden tongs, I add to 
 it the two bent pieces of wire x x (fig. 21) by means of 
 which the funnel is kept clofe to the mouth of the phial. 
 
 I throw
 
 I throw into it as much nitrous air as to be per- 
 fedly filled up with it: and I then replace this 
 phial b in the other mouth of the vefiel c (). '* 
 
 28. I take Afterwards the' Eudiometer with 
 rny left hand, holding it near the lower part 
 
 (h] No pains or trouble ought to be fpared, in order to 
 obtain, at any time, a nitrous air perfectly alike in its 
 contractive power, when mixed with common air. 
 
 In order to come the neareft to this, I take a phial D 
 (fig. 19,) like thofeyou have defcribed in the fecond volume 
 bf your work On different Kinds of Air : to the mouth of 
 which is ground air-tight the crooked tube n z in the fhape 
 of an S. I fill the half of this phial with thin brafs wire* 
 the thicknefs of which is equal to ^ of an Englifh inch, 
 nicely cut by a pin-maker to this length. I fill the three 
 quarters of the phial with common water ; and the re- 
 mainder with ftrong nitrous acid, which I have always 
 taken of the beft forr, at the Apothecary's Hall in Lon- 
 don. I put tne crooked tube z to the phial : and, as foon. 
 as the effervefcehce caufes the liquor to rife to the end z. of 
 the tube, I pafs it underwater into the mouth of the bottle 
 (fig. 20,) which is filled with water, and inverted with 
 its meuth downwards upon the hole of the fhelf n o 
 which appears covered with water within the trough, ot 
 pan, (fig. 17). 
 
 This figure reprefents the moft commodious fhape a 
 trough muft have for any experiments on different kinds 
 of air. It is made with ftrait boards of elm-wood one 
 inch thick. The infide dimenfibns are 25 inches longjj 
 13* wide, and 11 deep, Englifh meafure. The two end 
 iioards, c d and ef, are fitted into a groove cot in the other 
 E tbrc!
 
 <J, over the furface of the water in the trough, 
 to avoid breaking any of the phials, if it chan* 
 
 ces 
 
 three boards ; this is daubed with thick white painting, 
 as a cement, to keep well the water in : and the whole is 
 fattened with nails from the outfjde. The flielf nv a 
 is eight inches wide, and two inches thick. It has three 
 holes of three tenths of an inch diameter, wit!) as many fe- 
 parate cavities underneath, fo as to ferve like fo many fun- 
 nels. The figure, however, reprefents a glafs funnel / ce- 
 mented to the middle hole : which is equally convenient. 
 This ihelf is fuppcrted by four metallic hooks F<wxx t 
 which may be raifed or lowered at pleafure, by the wooden 
 wedges there reprefented. 
 
 When the bottle F is entirely filled by the nitrous air, I 
 ihot it up with its ftopple x (fig. 20,) which I pafs under 
 the furface of the water, to avoid any communication 
 with the external air: and I pufh this bottle under the 
 fhelf, where I let it remain for a quarter of an hour, to 
 acquire the fame temperature of the furrounding water : 
 and the fame I always obferve with the bottle, containing 
 that atmofpherical air which I defire to try, before I put 
 it into the phial b. 
 
 I muft acknowledge, however, that, notwithftanding thefg 
 precautions, I cannot fay that all the refults of my expe- 
 riments, even when made upon the fame atmofpherical air, 
 have as yet agreed fo exactly as I flattered myfelf they 
 would. Perhaps there was fome difference in the ftrength 
 of the nitrous air, the denfity of which I thought might 
 eafily be brought to a fettled ftandard,.to be determined 
 by means of a glafs hydrometer. Perhaps there was fome 
 ether little vantty in the circumftances of the experiments, 
 the influence of which 1 was not aware of. But let it be 
 as it way : I very willingly leave this problem to be re- 
 
 folvcd
 
 ccs to fall ; and, with my right hand, I turn the 
 veffel c upwards, fo that the two phials may 
 be downwards, as reprefented fig. 14. By 
 this operation the two kinds of air come up to 
 
 folved by abler chemifts than I can pretend to be : and I 
 heartily wifh they may fucceed better than I have done: 
 for, without being afTured of getting every where a cer- 
 tain flandard nitrous air, by which the fame atmofpherical 
 air be equally affedted, we cannot draw with certainty any 
 general decifive conclufions, from Eudiometrical experi- 
 ments made in diftant times or places. 
 
 Before I leave this fubjeft, I cannot help mentioning 
 two ftriking circumftances relating to nitrous air. The 
 firft is the great quantity produced by the action of nitrout 
 acid on many metals ; which may ftill be carried to a 
 greater extent, if helped by bringing the flame of a candle 
 to the phial, which contains the folution, when it fecms 
 to be nearly done with. The fecond is the antifeptic power 
 of nitrous air to preferve animal matters from corruption. 
 A beef-ftake, almoft entirely putrid, and with an infupport- 
 able ftench, being put into a jar of nitrous air, in lefs than 
 two days was perfectly reftored,and very eatable when drefl'ed. 
 A pigeon was very well preferved above fix weeks by the 
 fame treatment; and, when roafled, was found fo good as 
 to be eat without any diflike. Two other pigeons were kept 
 in it full fix months without corruption : they were ftill 
 very firm and of a good colour ; but the flefh had loft all 
 its flavour, and was far from being eatable when dreffed. 
 But the nitrous air for thefe ceconomical purpofes, which 
 may be of a great refource at fea, as well as at home, mufl 
 be made out of nitrous acid with iron, or other metal lefg 
 exceptionable than brafs or copper, the effluvia of which* 
 are pernicious to animals.
 
 ( 28 ) 
 
 y, from the phials a b ; and there they mix to- 
 gether in the beft poffible manner; the par- 
 ticles of each having a large room to come 
 into contact with each other ; fince the fore- 
 moft ones do not detain thofe which are be- 
 hind, as it happens when this mixture is made 
 in a narrow vefiel. This being done, I im- 
 mediately dip the Eudiometer in the water 
 of the trough (fig. 17,) leaving the mouch 
 of the instrument above its furface ; fo that no 
 more water may enter into it than what it h. d 
 at firft. I then obferve with attention the mo- 
 ment when the mixture x (fig. 14,) of the two 
 kinds of air comes to its greateft diminution ? 
 after which its bulk will begin to increaie 
 again. In order to catch this moment with 
 certainty, I fiide down the brafs ring z of the 
 instrument, as the furface of the water in the 
 tube falls. This point of the greateft dimi- 
 nution will be eafily perceived, by obferving 
 when that infide furface is ftationary : which 
 will happen in a few minutes, if the nitrous 
 air has a proper itrength (i). 
 
 29. As 
 
 (;') The bulk of the mixed air will decreafe to a certain 
 degree, within a few minutes, according to the ftrength of 
 nitrous air. Afterwards it will begin to expand again : 
 but this it will do to a very fhort limit, much below its 
 former bulk. This is a phenomenon which, I think, 
 I have ob/erved the firft on thele experiments ; hav-
 
 zg. As foon as the diminution of the twq 
 kinds of air appears to be frationary, I fill up 
 the whole tube of the Eudiometer with water : 
 I fhut it up with the ftopple m , and incline 
 the top of the inftrnment forwards, till the air 
 comes from x (fig. 14,) up to the top n of 
 the tube. I then keep the lower part of the 
 inflrument dipped in the water ; take off the 
 glafs veflel c with the two phials a b> and rife 
 or lower the tube of the Eudiometer, fo as to 
 fee the furface of the water, in the infide, even 
 with that in the outfide ; which I mark by 
 Hiding to it the brafs ring z. therwife I 
 apply the ruler (fig. n, without making any 
 ufe now of the brafs ring) to the fide of the 
 Eudiometer, whilfl it is immerfed in the water 
 of the trough : and there I fee the true di- 
 inenfion of the remaining bulk of the two 
 kinds of air, already diminimed. Perhaps the 
 beft method for this obfervation would be to 
 allow time enough that the mixed air may take 
 its fettled bulk : but this requires fometimes 
 twenty-four hours time. I leave, however, 
 the choice of thefe two methods to the obferv- 
 
 ing made a very great number of them with nice Eudio- 
 meters, of the kind I am now defcribing. It certainly 
 deferves the attention of Philofophers : and, although I 
 have communicated it to fome of my acquaintance, none 
 have as yet, in my humble opinion, given a fatisfa&ory 
 fblution of this phenomenon. 
 
 cr,
 
 ( 30 ) 
 
 pr, who may ufe both if he pleafes, provided 
 he keeps diftinctly the refult of each method 
 in his account of the experiment. 
 
 30. The number marked about the middle 
 of this ruler (fig. 1 i,) as for inftance, # #-96, 
 means that the contents of both phials a and 
 If are equal to ninety-fix divifions of the ruler, 
 when put into the tube of that Eudiometer : 
 that is to fay, they are equal to a folid cylin- 
 der, as thick as thp infide of the glafs tube, 
 and whofe length is ninety-fix divifions of the 
 ruler, which has been divided into tenths of 
 an Englilh inch,. 
 
 31. Now if, for inftance, this remaining 
 bulk of mixed air correfponds to the 56th di- 
 Tifion of the ruler, it (hews that, out of 9$ 
 parts, only 40 (=96 56) have been loft or 
 contracted : and, in this cale, the wholfome- 
 nefs of that air, which I call A, will be . 
 If another equal quantity of different air, which 
 I mail call B 9 had alfo been tried by the fame 
 Eudiometer, and its refiduum was equal to. 
 <>o parts of the fame ruler, the refpective 
 falubrity of the air B will then be to that of the 
 air A* as 36 (-9660) to 40. 
 
 32. But if the air B had been tried by ano- 
 ther Eudiometer, whofc proportional dimenfi-
 
 Oris, marked about the middle of its ruler^ 
 were # * = 108, then the refpedive falu- 
 brity of thefe two kinds of air A and J5, 
 would be in the compound ratio of -^ to i| 
 56x06 to 4ox 108 
 
 " 
 
 to 432 = 54 to 
 
 67, 5 : that is to fay, the wholfomenefs of the 
 air B would be to that of the air A 9 as 54 to 
 
 33. Nearly the fame refults would be found, 
 if the ruler (fig. 1 1,) was applied to the fide of 
 
 the 
 
 (I) It is fuppofed that the infide of the tube is of an uni- 
 form diameter ; but it often happens, that there are fomc 
 varieties in different parts of its whole length. When they 
 are not very confiderable, we may neglect their influence 
 in the refult of thefe Eudiometrical experiments ; but 
 when the contrary happens, it will be very eafy to make 
 a proper allowance for them in the calculation. It is for 
 this reafon that I have always ordered that the contents of 
 one fingle phial be marked alfo upon the fcale of each 
 Eudiometer, as well as the contents of both phials ; for 
 inftance as in. this manner: 
 
 * # = 96 
 
 * =47 
 
 which means, firft, that the contents of both phials 6 
 and b are equal to a cylinder, whofe diameter is the fame 
 as that of the infide bore of the tube d (fig. 16,) and 
 whofe height is equal to 96 equal divifions of the ruler : 
 fecondly, that the contents of a fingle phial are equal to 
 47 divifions in the upper part of the fame tube mud; and 
 
 of
 
 the Eudiometer, as foon as the incbfed mix- 
 ture of air came to its utmoft diminution, ai 
 mentioned No. 28 : becaufe as much water 
 muft fall in the tube n d^ as correfponds td 
 the diminution fuffered by the two mixed airg 
 in x. But there are fome varieties, which arife 
 from the different prefTure of the column of 
 water, which prefies more or lefs upon the air 
 .at x (fig. 14,) as it is longer or fhorter : and 
 thefe varieties ought not to -be overlooked iri 
 nice experiments : they are avoided by the 
 procefs already defcribed, No. 29 ; and majr 
 otherwife be prevented by the method of 
 which I will fpeak at the end of No. 39. 
 
 34. Whenever I have at hand a tall glafs 
 receiver, like that reprefented fig. 14, the 
 whole procefs is then more eafily performed ': 
 for in this cafe I dip the Eudiometer, inverted 
 as it appears fig. 12, into the water contained 
 in the veflel V S q I : I then put the two kinds 
 of air into the phials a and b as above faid, No. 
 26 and 27 : I turn the inftrument upright, as 
 reprefented fig. 1 4. -, and finifh the procefs, as I 
 have already defcribed. 
 
 of courfe, to 49 divifions (=196 47) of its lower part. 
 By this difference it appears that the tube of fuch Eudio- 
 ineter is wider in the top than at the bottom, by & of the 
 tthole. 
 
 35- t
 
 ( 33 ) 
 
 35- I mu ft> however, warn the operator 
 that, unlefs every trial, and even almofl every 
 part of the procefs, be made in the fame tempe- 
 rature ; or, at leafl, unlefs the varieties arif- 
 ing from this caufe be accounted for no re- 
 liance can be had on the refult of fuch expe- 
 riments : it being well known, that air is apt 
 to increafe or diminifh very considerably in 
 its bulk, by the influence of heat and cold. It 
 is for this reafon that I conftantly keep a 
 good thermometer K, which hangs by the 
 wire y r, and is immerfed in the water of the 
 glafs vefiel fig. 14, or in the trough fig. 17, 
 whenever I make- .any of thefe experiments. 
 For the fame reafon I take care to leave the 
 Eudiometer and the veflcls of ai r , immerfed in 
 water time enough, as above mentioned, to get 
 the fame temperature : and I make ufe of the 
 wooden tongs mentioned in note (), whenever 
 I handle the phials c b filled with air, chiefly if 
 they have not the folid lump at their bottoms, 
 as reprefented in the plate ; unlefs I feel the 
 heat of my hands to be the fame as that of 
 the water, in the trough, I make ufe of. 
 
 DESCRIPTION OF THE SECOND 
 NEW EUDIOMETER. 
 
 36. The Eudiometer, reprefented fig. 15, 
 
 confifts of a glafs tube / r, two or three feet 
 
 F long.
 
 ( 34 ). 
 
 long, and of an uniform diameter : the end c 
 is bent forwards ; and the other end / is wide 
 open, as a funnel, unlefs a feparate one is 
 made ufe of : this tube is faftetied, by two 
 loops, to the brafs fcale c w t V. There is 
 a glafs phial , the neck y of which Is ground 
 air-tight to the end / of the tube ; and contains 
 only half of the whole infide capacity of the 
 divided tube c t. It has, at the other end c, a 
 large round phial a b c, containing three or 
 four times the bulk of the phial n : its neck is 
 alfo ground air-tight to the mouth c of the 
 tube. The brafs fcale c w t Fis divided into 
 128 equal parts: this being a number that can 
 be divided to unity in a fubduplicate ratio 
 without fraction, by continual bilections -, on 
 which account it is one of the numbers the 
 late famous Mr. Bird had adopted for his di- 
 viding mathematical inftruments with the ut- 
 moft accuracy. Thefe numbers are fet out in 
 the fcale from / towards c. The contents or 
 capacity of the tube till the number 128 is the 
 double of the capacity of the phial n. Befides 
 this there is a tin veflel x s d t r o (fig. 15*}, 
 which may ferve as a packing cafe for the 
 whole inilrument, and its necefiary appen- 
 dages ; and alfo as a trough, when experi- 
 ments are made ; it being then filled with wa- 
 ter. Both the glafs tube reprefented fig. 22, 
 and the glafs floppier (fig, 15*), belong to 
 
 this
 
 ( 35 ) 
 
 jkis Eudiometer ; and both are fitted in, air- 
 tight, to its mouth V. 
 
 THE PROCESS. 
 
 37. Let the .inftrument be immerfed under 
 the water zz of the tin veflel fig. 15*: and 
 let the phial , filled with water, be put in the 
 infide focket e e d of the tin veflel. Let it be 
 filled with nitrous air, as above directed at the 
 end of No. 27 : and let this quantity of air be 
 thrown into the phial a b c (as directed No. 26 
 and No. 27), which I fix a little tight to the 
 mouth c of the Eudiometer. I afterwards fill 
 the fame phial n with the air I want to try : and, 
 raifing the end c of the inftrument, I put it into 
 its mouth V : when this is done, I fet the inftru- 
 ment upright, as reprefented fig. 15, hanging it 
 on the hook w ; and, as foon as this laft air goes 
 up to the phial a b c , I take off the phial n, 
 that the diminution of the two mixed airs may 
 be fupplied from the water in the tin veffel ; 
 which muft be the cafe, as the mouth ^"of the 
 Eudiometer is then under the furface of the 
 water. 
 
 38. I then put to the lower end V of the Eu- 
 diometer, the bent tube ftg. 22, to which is 
 fitted the brafs ring K, and is filled with 
 
 F 2 WW
 
 water. It is by obferving the furface of the wa- 
 ter in this fmall tube (which then forms a true 
 fiphon with the tube of the inftrumer.t) and 
 by means of the brafs ring X, that I can dif- 
 tinguilh the ftationary date of the diminifhing 
 bulk of the two mixed airs, above mentioned 
 at the end of No-. 28 : which being perceived, 
 I take off the fmall tube g h from the Eudio- 
 meter, and lay down, for fome minutes, the 
 whole inftrument, in an horizontal pofition, 
 under the water of the tin veflel : I mut up 
 the mouth V with the glafe ftopple m , and, 
 reverfing the inftrument, I hang it up by the 
 end V* on the hook w. By this pofition 
 the whole diminiffced air of the veflel a b c goe$ 
 up to the top, where its real bulk is fhewn by 
 the number of the fcale, facing the infide fur- 
 face of water. This number being deducted 
 from 128, gives the comparative wholfome- 
 nefs of the. air already tried, without any fur- 
 ther calculation. 
 
 39. Biitthis procefs will be ftill eafier, when 
 the laft 'diminution of the two mixed kinds of 
 air, mentioned No. 29, is only required in the 
 obfervation : becaufe no ufe will be then made 
 of the fyphon, (fig. 22). In fuch a cafe the in- 
 ftrument is left hanging, on the hook wfor 48 
 hours: after which it is laid down under the 
 water of the trough (fig. 15*), in an horizon- 
 tal
 
 ( 37 ) 
 
 / 
 
 tal pofition, for 8 or 12 minutes, in order to 
 acquire the fame temperature of the water : 
 the mouth F"is then fhut up with the ftopple m; 
 the inftrument is hung by the end V in a con- 
 trary pofition, and the Jail real bulk of the good 
 mixed air will be then fhown by the number 
 of the brafs fcale anfwering; to the in fide fur- 
 face of the water. This number being fub- 
 tracted from 128, will give the comparative fa- 
 lubrity of the air employed in the trial, without 
 any further calculation. I need not fay that all 
 the circumftances already mentioned for the bet- 
 ter obtaining exact refulcs in thefe experiments, 
 are to be carefully obferved, when this fecond, 
 or the third Eudiometers areufed: but chiefly 
 that circumftance, mentioned No. 35, ought 
 never to be omitted. The thermometer is to 
 be kept dipped in the water of the tin vefiel ; 
 and the Eudiometer mult be kept there im~ 
 merfed fome minutes, as I have faid juft now, 
 before it is railed for the laft time, to read off 
 the quantity of the total diminution of the mix- 
 ed air. As to the other circumftance, menti- 
 oned No. 33, it has been rendered unneceflary 
 by laying this Eudiometer in an horizontal 
 pofition before the glafs ftopple was put in. 
 The fame method muft be applied to the third 
 new Eudiometer I am going to defcribe ; and 
 even the firft Eudiometer, already defcribed, 
 may be treated in the fame manner : for if it be 
 
 laid
 
 ( 38 ) 
 
 laid down in an horizontal pofition under the 
 water in the tub, before it be (hut up with 
 the ftopple, as directed No. 29, there) will be 
 no variation produced by the expanfion of the 
 air in the infide : becaufe the proper quantity of 
 water is then fhut up within the glafs veffd c 
 of the inftrument: fo that raifing it up, as it is, 
 together with the veflel c, and its phials a I (fig. 
 1 4), the weight of the column of water will prefs 
 totally upon them, without expanding the in- 
 clofed air, or caufing any variation beyond the 
 trifling one which may proceed from the natu- 
 ral elafticity of the fides of the glafs tube and 
 veflels. 
 
 40. I muft, however, acknowledge that, the 
 long way through which the air pafTes, in going 
 at firft to the large phial a b c in this fecond 
 Eudiometer, muft leave fome doubt whether 
 it has not then fuffered fome fenfible change 
 in its quality, before it is mixed with the ni- 
 trous air ; fince, as you have obferved, the 
 air that has been long agitated in water, changes 
 for the better from its bad qualities : and this 
 objection muft be ftill greater in the ufe of the 
 third Eudiometer. It is on this account that I 
 have mentioned the firft Eudiometer, as the 
 leaft exceptionable of all that we know till 
 the prefent ; and perhaps the nature of the 
 thing is not capable of a further perfection. 
 
 Indeed
 
 39 ) 
 
 Indeed that inftrument, I mean my firft Eu- 
 diometer, has not only the advantage of offer- 
 ing a very fmall way through the water to the 
 two kinds of air, on their going to mix at x in 
 theveffcl c (fig. 14), but they are kept fepa- 
 rated till that moment, in the two refpective 
 phials a and , without any other contact with 
 the water, but only in the narrow diameter of 
 the necks of thefe phials. 
 
 DESCRIPTION OF THE THIRD NEW 
 EUDIOMETER. 
 
 41. This third Eudiometer is the neareft to 
 your original one: and, was it not for the con- 
 fideration I have mentioned in the preceding 
 number (which, perhaps, will not weigh too 
 much with fome philofophers) and few other 
 circumflances which are obvious, I mould not 
 doubt to pronounce this third inftrument to be 
 the beft of all the three, as I have advanced in 
 my laft letter to you of the 3Oth of November 
 laft. This, however, I gladly fubmit to your 
 fuperior judgment. 
 
 42. This third Eudiometer confifts- of a 
 ftrait glafs tube e n (fig. 8), of an uniform diar 
 meter, about two or three feet long, with a large 
 ball j, and a glafs ftopple m, fitted air-tight to 
 
 the
 
 ( 40 ) 
 
 the mouth #, which ought to be wide open, 
 as a funnel, unlefs a feparate one is made 
 ufe of. There is alfo a fmall fiphon (fig. 23) 
 with a brafs ring x: a fmall phial z (fig. 
 9), the contents of which may be received in 
 the third part of the ball s : and, when put 
 into the glafs tube n j, muft take there no more 
 than the half of its length. ' Laftly, this in- 
 liniment has a ruler (-fig. 13), which is divided 
 and ftamped like that other already defcribed 
 at the end of No. 24 ; and a glafs funnel, which 
 is ground to the mouth n of the inftrument, 
 when this is not wide t>pcn, as already faid. 
 
 THE PROCESS. 
 
 43. The ufe of this inftrument is eafily un* 
 derftood by what I have already faid of the tw* 
 preceding ones. 
 
 Firft, it is filled with water, and fet in a 
 vertical pofition, with the mouth n under the 
 furface of the water in a tub, or in a trough, 
 
 Secondly, the phial 2 (fig. 9) is filled, as 
 
 above, with nitrous air- 9 and thrown into the 
 
 tube by means of the glafs funnel y (fig. 10), 
 
 larhich is ground to the mouth of the Eudio- 
 
 g meter,
 
 meter ; unlefs it be wide enough not to be in 
 need of any funnel. 
 
 Thirdly, the fame phial z is again filled 
 With the air to be tried ; and thrown into the 
 fame. 
 
 Fourthly, .the fiphon (fig. 23) is added im- 
 mediately to the mouth n of the Eudiometer, 
 under the furface of the water ; fome of 
 which is to be poured into it. 
 
 Fifthly, the ftationary moment of the great- 
 eft diminution of the mixed air at j,. is watched 
 by means of the ring x, as mentioned No. 28 
 and 38. 
 
 Sixthly, when that moment arrives, the fi- 
 phon K I (fig. 23) is taken off; the Eudiome- 
 ter is laid for fome minutes under the water, 
 in an horizontal pofition, or nearly fo, but in 
 fuch a manner that no part of the inclofed air 
 may get out ; the rnouth n is mut up with the 
 glafs ftopple , and theinftrument is inverted 
 with the mouth n Upwards. 
 
 Laflly, the fpace occupied by the refiduum.- 
 of the diminimed air, is meafured by applying 
 to its fide the divided ruler, or fcale (fig. 13), 
 and the refuk is efnmated after the manner al- 
 ready explained No. 31 and 32. 
 
 G 44>
 
 ( 42 ) 
 
 44- Whenever I want only to know the laft 
 diminution of the mixed air, mentioned No. 
 39, the procefs then becomes eafier, as no ufe 
 is made of the fiphon (fig. 23). The method 
 of conducting the procefs in fuch a cafe be- 
 ing refpectively the fame as that already de- 
 fcribed No. 39, it isunneceflary to defcribe it 
 here again. The fame precautions I have fpo- 
 ken of. No. 35 and 39, muft be obferved when 
 this Eudiometer is made ufe of, in order to form 
 a true judgment concerning thofe places, where 
 people will be able to live without danger of 
 hurting their conilitutions, by breathing and 
 being continually 'furrounded by noxious air; 
 which they have not yet been able to diftin- 
 guilh from the moft wholfome, except by a 
 long and too late experience. 
 
 45. The Eudiometers already defcribed are 
 the fiteft inftruments for philofophical experi- 
 ments, on the bulk of air and other fluids, 
 when mixed together ; and even when mixed 
 vith fome folid fubftances, which can be in- 
 troduced into the lower veflel c of the firit 
 of the three Eudiometers. It will be better, 
 however, to have them made purpofely for 
 fuch obje&s, with a tube two or three times 
 longer than I have indicated above. When- 
 ever dephlogifticated air is to be tried by thefe 
 inftruments, proper care is to be taken to ob- 
 ferve the precife point of its full faturation, 
 3 which
 
 ( 43 ) 
 
 which is that of its greateft diminution by the 
 addition of nitrous air. 
 
 46. In order to make this experiment with 
 great accuracy, let a narrow glafs tube of an 
 uniform diameter (fig. 24), be provided : let 
 one of the two phials a or b (fig. 1 6) filled with 
 qnickfilver, be thrown into it, and the tube 
 cut exactly to thatfize, fo as to contain neither 
 more nor lefs. Let its whole length be divided 
 into fome number of equal parts, by which num- 
 ber the value marked on the ruler (fig. u), of 
 this Eudiometer, can be divided without any 
 fraction : for inftance, the number % % ~ 
 108, marked in the ruler, means, that the 
 contents of the two phials a and , of which 
 I fpoke No. 32, are equal to a cylinder of 
 1 08 divifions long, as thofe of the ruler : and, 
 of courfe, it mews that a fingle phial a or b 
 contains but 54 of thefe parts. Jn this cafe 
 this tube (fig. 24) may be divided either into 
 27 parts, each containing two of the ruler -, 
 or into 54, into 108, &c. 
 
 N. B. If the top of the tube is not very flat 
 in the infide, it will be more exact to divide 
 the weight of the quickfilver in two parts ; 
 to put one of them into the tube ; to mark 
 the fpace occupied by it ; to divide the pare 
 pf it, which was empty, into half th? number 
 G z i-
 
 ( 44 ) 
 
 intended for this tube, and afterwards to di- 
 vide the other half into fimilar equal parrs, as 
 the firft half, carrying them towards the clofed 
 end. 
 
 47. If the dephlogifticated air is very pure, 
 it will require almoft double the quantity of 
 nitrous air to be completely faturatcd. In orr 
 der to do this without exceeding the neceflary 
 quantity, I throw into the tube nd fig. 17) 
 a fecond meafure b or a of nitrous rf/r, after I 
 have brought the procefs to the moment men- 
 tioned No. 29 : in this cafe the whole volume 
 or bulk of the dephlogifticated and nitrous air 
 will be 162 [=108 + 54:] I obferve where the 
 furface of the infide water in the tube ftops, 
 and I mark it by .the fliding brafs ring z. I 
 then fill up the divided tube (fig. 24) with 
 nitrous air : I throw a fmall quantity into the 
 Eudiometer tube n d ; and, if it becomes of a 
 redifh colour, the inclofed air will diminifh : 
 I then pum up the ring z, and, by this means, 
 I go on throwing in the nitrous air, by little 
 and little, till I fee that the whole diminimes 
 no more ; which mews me that it is fully fa- 
 turated. 
 
 48. Let us fuppofe, for example, that the 
 tube (fig. 24 ) was divided only into 27 equal 
 parts i and that the faturation of the dephlo- 
 gifticated
 
 ( 45 ) 
 
 gifticated air was compleated at the eighth di- 
 vifion of it : this fhews that 19 parts [27 8 
 19], equal to 38 of thofe marked in the ru- 
 ler,- have been thrown into the Eudiometer ; 
 that is to fay, that the whole bulk of both 
 kinds of air is equal to 200 [ 162 + 38] 
 meafures, as thofe marked by the ruler (rig. 
 ii,) already explained No. 30. Now if the 
 remaining quantity of air within the Eudio- 
 metrical tube is only equal to two meafures or 
 numbers of the ruler, it is clear that fuch de- 
 phlogifticated air is ninety-nine times of an 
 
 hundred rlHi = ii 8 = 39-1 pure air; 
 
 L 200 200 IOO J 
 
 fince its bulk is reduced, by the combination 
 of nitrous air, to the -^-^ of the whole. 
 
 49. It is but three days ago (/) that you 
 (hewed me fuch a wonderful kind of air, as I 
 have exemplified in the preceding number. 
 This air you have produced before my eyes, 
 from a folution of quickfiher and nitrous acid 9 
 
 (/) This additional article to the prefent letter was wrote 
 on the i6th of September, 1777; although the greateft 
 part of it has been written many months before, and the 
 firft twenty numbers were already printed : but fome cir- 
 cumftances, the knowledge of which cannot intereft the 
 public, have hindered the publication of the whole tilfe 
 this prefent time. 
 
 made
 
 ( 46 ) 
 
 made many months before, and then diftilled 
 in a long but narrow glafs retort, with a fand- 
 heat. This is, perhaps, inleed, an extraor- 
 dinary phenomenon, and feems to bring us a 
 little nearer to the door of the fecret laboratory 
 of Nature in the formation of air. 
 
 50. I cannot fay, but fo pure a dephlogijlica- 
 ted air may flill be produced by this procefs ; 
 that its whole bulk may be reduced to nothing 
 by a proper combination with nitrous air. If 
 fo, what fhali we then be able to think of a 
 fluid fubftance, which is coercible in a glafs 
 veflel, to which above the double quantity of 
 another fubftance [VV 6 2 >7] likewife coerci- 
 ble in a glafs veffel, being added j both thefe 
 fubftances to appearance wholly vanifh 1 
 
 I JO" 7 ?' 
 
 51. This phenomenon certainly deferves the 
 attention of philofophers : and I gladly leave 
 to them the examination of it. I muft only 
 add, for their information, that the nitrous 
 acid is the thing chiefly concerned in its pro- 
 duction : when this admirable fubftance acts 
 on certain kinds of bodies, as quickfiher in the 
 prefent cafe, its folution produces that elaftic^ 
 but coercible fluid, which we call nitrous air : 
 the refiduum, after a long while, being pro- 
 perly urged by fire, gives at laft the other 
 
 ici but likewife coercible fluid, which we 
 
 call
 
 ( 47 . ) 
 
 call depllogifticated air 5 and the combination 
 of both, nearly in the above proportion, pro- 
 duces the wonderful phenomenon I have fpo- 
 ken of. 
 
 52. I mall fay no more on this matter; and 
 leave it very willingly to be confidered and un- 
 ravelled by abler philofophers than I can pre- 
 tend to be : and conclude the fubject of this 
 letter by alluring you, that I mall be very 
 happy, if the things here treated of mould 
 deferve your approbation : and ilill more fo, 
 if they produce the defired effect I aim at, 
 the general good of mankind. I am, with 
 the utmoft regard and fmcere friendfhip, 
 
 My dear Sir, 
 
 Your moft obedient and 
 Affectionate fervant, 
 
 Bowood Park, j R ^ MAGELLAN. 
 
 Januarys, 1777.
 
 - r
 
 Tf. ft* ^^^L & 
 
 J3 =? 
 
 % I/ 
 
 I a 
 
 10S-ANGEI& 
 
 -UBRARY 
 
 
 I 
 
 I 8 
 
 S s 
 
 
 g I 
 
 
 1 f 
 
 ^ > 
 
 I I 
 
 .^lOS ANGELA 
 ^ t>^-^^ 
 
 <-S ^- 
 
 I I 
 
 It 
 
 -n V 
 
 f fc 
 
 ^ g 
 I 1 
 
 ^W 
 
 IW3-JO^ 
 
 
 
 ^avuan-^
 
 3 
 
 
 wnw 4>mmw ^fliwsov 
 
 I I 
 
 ^1 \ 
 
 .% 
 
 ft 3' 
 
 
 1 1 
 
 
 I 1 
 
 vsov