THE QUARTERLY JOURNAL OF TWE CHEMICAL SOCIETY. Nov. 4 1850. THOMAS GRAHAM,EsQ. V.P. in the Chair. The following Donations to the Library have been made since the last Meeting ‘‘ The Pharmaceutical Journal,” Vol. X. Nos. 1 2 3 4 and 5 by the Editor. “ Journal of the Franklin Institute,” Vol. XIX. Nos. 5 and 6; Vol. XX. Nos. 1 and 2 by the Institute. “Transactions of the Iloyal Society of Edinburgh,” Vol. IV- XIX ; Part 1 Vol. XX ;and the i‘Proceedings of the Royal Society of Edinburgh,” Parts 35-39 ; by the Society. ‘‘ On the Constitution of Codeine and its Products of Decomposi-tion,” by Thomas Andrews M.D. by the Author. “Report to General Sir Thomas Macdougall Brisbane Bart. on the completion of the publication in the Transactions of the Royal Society of Edinburgh of the observations made in his Observatory at Markerstown,” by John Allen Brown presented by the Author.“On the Isomorphism and Atomic Volume of some Minerals,” by James D,Dana by the Author. Berichte iiber die Mittheilungen von Freunden der Naturwissen- schaften in Wien,” von Wilhelm Haidinger. Band V. and VI. 1848-9. rr Bulletin of the Royal Belgian Academy of Medicine,” Vol. IX? Nos. 3-7 by the Academy. CCQuarterly Journal of the Geological Society,” Vol. VI. No. 23 by the Society. “ Setzungsberichte der Kaiserlichen Akadernie der Wissenschaften Mathematisch -Naturwissenschaftliche Classe.” Erste Abtheilung (Jan. Feb. March April). “Haidinger’s Report,” Vol. V and VI. 1849-50 by the Society of the Friends of Natural Science in Vienna.“ Abhandlungen der Mathemati~ch-physikalischenClasse der KO-niglich Bayerischen Akademie der Wissenschaften.” Band V. 1 2,3 Abtheilungen. VOL. 111.-NO. XII. Y DR. FRANKLAND’S RESEARCHES I‘ Chemical Investigation of the most important Mineral Waters of the Duchy of Nassau,” by Dr. R. Fresenius by the Author. “ Ueber den Antheil der Pharmacie an der Entwickelung der Che- mie :, von Dr. Ludwig Buchner Jun. “Laurent and Gerhardt’s Comptes Rendus,” (from Quesne- ville’s “Revue Scientifique”) Nos. 1-9 by M. Quesneville. “Taylor’s Calendar of the Meetings of the Scientific Bodies of London for 1850-1” by the Publisher. “Silliman’s American Journal for September 1850” by the Editor. (‘Quesneville’s Revue Scientifique from December 1849 to May 1850 by M.Quesneville. “On the Power of Soils to absorb Manure,” by Professor Way by the Author. “ De Cerevisiae Vera Mixtione et indole Chemica et de Methodo analytica alcoholis quantitatem recte explorandi,” scripsit Dr. H. Wackenroder by the Author. The following Papers were read cr On the Magnetic Attraction of Metals,” by Mr. Richard Adie Liverpool. XXVI1.-Researches on the Organic Radicals. BY E. FRANKLAND, PH.D. F.C.S. 111. ON THE ACTION OF SOLAR LIGHT UPON IODIDE OF ETHYL. The action of light in modifying and controlling chemical affinity has frequently been the subject of investigation and the exceedingly curious and interesting results which have already been observed seem to promise that this agent will become a most valuable assistant in chemical research.Gay-Lussac and Thenard were the first to point out that chlorine and hydrogen may be preserved in contact for any length of time without entering into combination if the mixture be carefully preserved from light; but that with the presence of light combination immediately commences and proceeds with a rapidity proportional to the luminous influence. According to Faraday iodine and olefiant gas combine most readily in sun-shine; and the name phosgene gas was given to chlorocarbonic acid because light was found to be essentially necessary to its formation. These are a few of the instances in which light has been observed to produce direct combination ; but the cases in which it effects decom- position and changes the order of elective affinity are much more numerous.Under the influence of light chlorine is enabled to decompose water uniting with its hydrogen and liberating pure oxygen gas and according to Grotthuss the blue solution of iodide of starch ON THE ORGANIC RADICALS. 323 in water is completely decolorized with the production of hydriodic acid. Scheele Seebeck and others found that nitric acid exposed to sunlight is converted into nitrous acid and oxygen whilst many metallic oxides lose the whole or a part of their oxygen thus peroxide of lead is resolved into minium and oxygen ; grey oxide of mercury into metallic mercury and red oxide; whilst red oxide of mercury under water is decomposed into grey oxide and oxygen gas.Oxide of silver is resolved into silver and oxygen; carbonate of silver into silver oxygen and carbonic acid; and oxide of gold into gold and free oxygen Boullay finds that aqueous solution of perchloride of mercury is decomposed into protochloride of mercury hydrochloric acid and oxygen gas. In addition to these curious reactions the highly interesting and important discoveries of Hunt Daguerre Herschell and Talbot need only be men- tioned to establish the great scientific as well as practical importance of this remarkable function of light which is also so completely under the control of the operator admitting of being increased diminished or modified at pleasure that there seems every proba- bility of this agent becoming one of our most valuable means of composing decomposing and ascertaining the rational constitution of organic bodies especially as it allows of being applied in such a convenient manner and under circumstances in which other means are inapplicable.It has been long known that certain inorganic bodies containing iodine such for instance as the iodides of silver and gold undergo decomposition when exposed to light the iodine compounds of the noble metals appearing to be most susceptible of this change. From the close relation of hydrogen to these metals its iodide might be expected to possess the same susceptibility and this is in fact found to be the case; for it is well known that aqueous hydriodic acid even when preserved in closely stopped bottles gradually turns brown on exposure to light from the separation of free iodine but the decom- position only becomes continuous when the iodine is removed as fast as it is liberated; it has also been observed that when hydriodic acid gas is allowed to stand over mercury its volume becomes reduced to one half and the residual gas consists of pure hydrogen; but whether this reaction only occurs under the influence of light has not yet been clearly established.In a former memoir,* I pointed out the very close analogy existing between the functions of the compound alcohol-radicals and those of the simple radical hydrogen as exhibited firstly in the constitution * Chem. SOC. Qu. J. 111 47. Y2 DR. FRANKLAND’S RESEARCHES of the vapours of their parallel compounds; secondly in the decom- position of their respective iodides by zinc; and thirdly in the replacement of hydrogen by methyl cthyl zlrnyl &c.in cacodyl in the new nitrogenous bases of Wurtz and Hofmann and in those of M. Paul Th6nard coutaining phosphorus. In order to add another link to the chain of evidence which already exists in support of this analogy it appeared interesting to ascertain if the action of light upon the iodides of ethyl methyl &c. would yield results corresponding to those obtained from iodide of hydrogen under similar circumstances; and this link seemed to me also of greater importance on account of the results thus produced by the simple action of an imponderable agent being less likely to be influenced by the formation of secondary compounds than in the action of zinc upon these bodies at an elevated temperature.My experiments have as yet been principally confined to the iodide of ethyl and it is the results of the action of light upon this body that I have now the honour to communicate to the Society. It has been remarked by almost all chemists who have had occa- sion to employ iodide of ethyl that this liquid slowly becomes brown from the separation of iodine when exposed even to diffused day- light ;this observation which I have myself of late also frequently had an opportunity of making induced me to hope that a decom-position here occurs analogous to that suffered by iodide of hydrogen under the same influence.I find that the ethyl-compound when exposed to direct solar light rapidly becomes of a dark-brown colour; but as is the case with hydriodic acid this separation of iodine soon ceases and when a certain intensity of colour has been attained no further action takes place; if however the free iodine be removed by agitating the liquid with mercury the action immediately recom-mences and proceeds to the same point as before. This behaviour of the iodide under the influence of light and in contact with mer-cury indicated the method by which the action could be carried on continuously and the products collected and preserved. For this purpose several glass flasks of about 10 ounces capacity were filled with mercury and inverted in a vessel containing the same metal; a few drops of iodide of ethyl being then introduced into each by means of a pipette they were exposed to the direct rays of the sun.The surface of the mercury where it was in contact with the liquid soon became covered with a film of protoiodide which by the further action of the light was converted into binio- dide whilst bubbles of gas were continually evolved and gradually displaced the mercury from the flask finally the whole of the iodide ON THE ORGANIC RADICALS. of ethyl disappeared the gas and biniodide of mercury being the sole products of the decomposition. Although simple exposure to the sun’s rays caused this action to take place with tolerable rapidity yet it was very greatly accelerated by placing each flask near the focus of an 18-inch parabolic reflector which was not however SO highly polished as to cause a very considerable elevation of tempera- ture the heat never rising to the boiling-point of iodide of ethyl (71.6’ C.) in this manner a few hours’ exposure sufficed to fill the flasks with gas which was then transferred to the bell-glass figured in a former Memoir,* and allowed to stand over sulphuretted water for twelve hours.At the end of this time all traces of iodide of ethyl vapour had been absorbed and the gas was fit for the endiome- trical operations. As iodide of ethyl is not in the least acted upon by mercury at a temperature of 150’ C.,? it could scarcely be supposed that the com- paratively low degree of heat to which these materials were exposed in the focus of the reflector could play any importaut part in the decomposition; yet in order to set this question entirely at rest an inverted bell-jar containing iodide of ethyl confined over mercury was surrounded by a glass cylinder and this latter filled first with water then with a solution of chloride of copper and lastly with a solution of bichromate of potash.When the outer cylinder W&S filled with water the decomposition proceeded with as much rapidity as without the intervention of that fluid whilst the temperature of the water was scarcely perceptibly raised during the operation ;the same was the case when solution of chloride of copper was employed ; but on substituting the solution of bichromate of potash scarcely the slightest action was perceptible even after several days’ exposure to bright sunshine.Now since according to Mr. Hunt at whose suggestion I employed these liquids the solution of chloride of copper absorbs nearly all the heating rays and allows about 90 per cent of the actinic rays to pass whilst the solution of bichromate of potash intercepts the actinic and gives free passage to the heating rays it is evident that the decomposition before us is due to the chemical influence of light and is totally independent of the heating rays of the solar spectrum. The gas collected and preserved as just stated was then submitted to the eudiometrical processes minutely described in my former Memoir on the cc Isolation of Ethyl,”$ the observations were also made with similar precautions respecting temperature pressure and formation of nitric acid during the explosion with excess of oxygen.* Chem. SOC. Qu. J. 11 266. j-Chern. SOC. Qu. J. 11,295. ? Chem. SOC. Qu. J. 11 269. DR. FRANKLAND’S RESEARCHES The entrance of atmospheric nitrogen into the gas during the time occupied in its purification from iodide of ethyl vapour by standing over sulphuretted water was sought to be avoided by using a deep bell-jar of comparatively small diameter and an outer glass cylinder,* having its internal diameter only 8 inch greater than the external diameter of the interior jar; thus a very small surface of liquid was exposed to the atmosphere and therefore only a proportionably small amount of nitrogen could be absorbed and transferred to the internal gas.The experiments detailed below show that the plan was effectual as none of the gases examined contained an appreciable amount of that element. A determination of the specific gravity of the gas gave the fol-lowing numbers Temperature of room . 2403~ C. Height of barometer . 767-6 mm. Height of inner column of mercury . 13-6 , Weight of flask and gas . 31.6758 grms. 25*3OC. Temperature in balance-case Weight of flask filled with dry air -31.5559 grms. Temperature in balance-case . 25-7O C. Capacity of flask . 140.79 cbc. From these data the specific gravity was calculated to be 1.7159. To ascertain the composition of the gas it was first subjected to the action of fuming sulphuric acid ; two specimens were then exploded with atmospheric air and excess of oxygen and the residue was treated with recently boiled absolute alcohol.The following numbers were obtained In absorption eudiometer. 1. Difference of Correctedvol. Observed Temp. mercury Baromr. at 00 C. and vol. level. lmpressure. Vol. of gas used )175,4 24*4OC. 603~~ 761*Omm121.51 (dry)* Vola. after action oft 150.6 19~5~ 19.0 , 759.5 , 104-08 8% (dry)= , Vol. after removal of specimen for 19*8O, 62.1 , 766.3 , 57.05 combustion. Vol. after absorp-) o.o tion by alcohol. * See Fig. 2. BB Chem. SOC. Qu. J. 11 266. ON THE ORGANIC RADICALS. 327 In combustion eudiometer. 11. Difference of Correctedvol. Observed Temp.mercury Barom'. at Oo C. and vol. level. Im pressure. Val* of gas used } 94.8 17.40c. 601.0- 7643mm 13% (moist). Vol. after admission 472.2 17.70,, of air (moist). 1 Vol. after admission 567,() 17.90,, of 0 (moist). 1 Vol. after explosion (moist). } 525.7 18.10, '01* after abioqtion 467.8 16.@, of CO (dry). } after admission } 683.3 I?'@,, of H (dry). 20093 , 7644, 24345 112.8, 763.9, 338.30 150.5, 763.6, 29466 2041.7, 765.2, 247.34 11.9, 765.7 , 48436 Difference of Correctedvol. mercury Baroxnr. at 00 C.and level. lmpressure. III. Observed Temp, vol. vol* Of gas used (moist). VOL after admission of air (moist). Vol. after admission of 0 (moist). 98.2 17.30C. 595.4mm 7649mm 1429 511.9 17.60 ,, 1 602.9 18.10,, I.161.5 , 7647 , 282.85 85.1 , 7648, 375.53 123.6, 765.3 , 328.57 176.2, 7644 277.90 '01* after explOsion 559.8 18.20 ,, (moist). 1 Vol. a.fter absorp-1501.9 17*0°,, tion of CO (dry). Analysis No. I. proves that the gas is a mixture and contains in 100 parts Gas absorbable by SO . . 14.34 Gas unabsorbable by SO . . 85.66 100.00 And the perfect disappearance of the gas left unabsorbed by sulphurio acid on subsequent treatment with about an equal volume of absolute alcohol proves the absence of hydrogen and hydride of methyl (light DR. PRANKLAND'S RESEARCHES carburetted hydrogen) as also any appreciable amount of nitrogen which might have permeated the sulphuretted water used to confine it during purification.Analysis No. 11 shows that 13.24 vols. of the gas remaining after the action of fuming sulphuric acid consumed 7'7.72 vols. oxygen and generated 47.32 vols. carbonic acid causing a con-traction on explosion equal to 43.64 vols. According to analysis No. 111. 14-29vols. combustible gas con- sumed 83.34 vols. oxygen and generated 50.67 vols. carbonic acid causing a contraction of 46.96 vols. If we take into consideration the composition of the gases evolved by the action of zinc upon iodide of ethyl,* together with the proof given below that the gas absorbed by sulphuric acid has exactly the composition and state of condensation of olefiant gas there can scarcely be a doubt that the gaseous mixture remaining after the action of sulphuric acid consists of ethyl and hydride of ethyl; and by forming two equations in which the volume of the mixture and the amount of contraction produced by explosion with oxygen are taken into account it is easy to ascertain their respective volumes even independently of the quantity of oxygen consumed and car- bonic acid generated although these latter values may be used to control the result arrived at by employing the amount of contraction only in the calculation.I prefer employing the observed contraction as the known quantity in the second equation to either the volume of oxygen consumed or that of carbonic acid generated because the number representing it is obtained from two readings which are least liable to slight sources of error and contains within itself the results of the entire analysis viz the volume of combustible gas the oxygen consumed and the carbonic acid generated.It will be obvious on inspecting its formula that 1 vol. of ethyl requires for its combustion 6.5 vols. oxygen and generates 4 ~01s. carbonic acid; and as the contraction which occurs on explosion is equal to the volume of the combustible gas + the volume of oxygen consumed -the amount of carbonic acid generated it is evident that the gas in question will cause a con- traction equal to 3.5 times its own volume; arid for similar reasons- since hydride of ethyl consumes 3.5 times its own volume of oxygen and generates twice its volume of carbonic acid-this gas must produce a contraction on explosion equal to 2.5 times its own volume.* Chem. SOC. Qn. J. 11 277. ON THE ORGANIC RADICALS. If then we represent the volume of this combustible mixture by A the contraction produced by explosion with excess of oxygen by B and the volumes of ethyl and hydride of ethyl respectively by x and y we have the following equations x+ y=A 2x+-y=B 75 2 from which the values of x and y are found to be 2B-5A X= 2 7A-2B Y= 2 and by substituting the numbers found in analyses Nos. 11. and III. for A and B we have 11. 111. LI = 10.54 11*23 y = 2.70 3.06 13-24 14.29 Hence the gas unabsorbed by fuming sulphuric acid contains in 100 parts XI. 111. MEAN. Ethyl Hydride . of Ethyl . . 79.61 20.39 78.59 21.41 79.10 2090 100*00 100*00 100*00 In order to ascertain the composition and state of condensation of the gaseous body absorbed by fuming sulphuric acid the original gas before being exposed to the action of that acid was exploded with atmospheric air and excess of oxygen.The following results were obtained IV. Difference of Correctedvof. Observed Temp. mercury Baromr. at 00C. and vol. level. lm pressure. Vol. of gas used 1 96.3 (moist). 9*4OC. 575.9"" 747.5"" 15.16 VOLafter admission 1590.r 9.5*, 68.9, 746.7 , 381*83 of air and 0 (moist) Vol after explosion } 549.2 (moist). 9*7O> 105.4, 746.7 , 335.34 Vol. after absorp- p92.1 tion of CO (dry). 7*4*, 158.1 , 753.1 , 28459 DR. FRANKLAND’S RESEARCHES According to this analysis 15.16 vols.of the gas consumed 82.08 vols. oxygen and generated 50.75 vols. carbonic acid; but analysis No. I. shows that this quantity of gas contains 12.99 vols. of the mixture of ethyl and hydride of ethyl which according to the mean of analyses Nos. 11. and III. would consume 7599 vols. oxygen and generate 46.24 vols. carbonic acid; thus leaving 6-09vols. oxygen and 4.51 vols. carbonic acid as the oxygen con-sumed and carbonic acid generated by the 2.17 vols. of the gas absorbable by sulphuric acid 1 vol. of which must therefore have consumed 2.81 vols. 0 and generated 2-08vols. carbonic acid Vol. of comb. gas. 0 cmsumed. CO generated. 2.17 6.09 451 1 2-81 2-08 When we reflect that in this experiment all the errors of observa-tion are concentrated upon a very small proportion of the gas sub- mitted to analysis the numbers obtained agree sufficiently with those yielded by the combustion of olefiant gas to allow safely of the conclusion that the body absorbed by sulphuric acid is the gas in question; for 1 vol.of olefiant gas consumes 3 vols. oxygen and generates 2 vols. carbonic acid. The composition of the gases evolved by the action of light upon iodide of ethyl in presence of mercury may therefore accordingIto the mean of the above analyses be thus centessimally expressed Ethyl . . 67.76 Hydride of ethyl . . 17.90 Olefiant gas . . . 14.34 100*00 The theoretical specific gravity of 8 gaseous mixture of this composition agrees closely with that found by experimenty as shown by the following calculation C H = 67.76 x 2.00390 = 135.7843 C H, H = 17.90 x 1.03652 = 18.5537 C H = 14.34 x 0.96742 = 13.8728 100~00 168.2108 a = 1.682108 100 Specific gravity as found by experiment .. 1.7159 Hence the decomposition suffered by iodide of ethyly under the ON THE ORGANIC RADICALS. influence of light and in presence of mercury is expressed by the following equation a small portion of the liberated ethyl being at the same time trans- formed into equal volumes of olefiant gas and hydride of ethyl The slight deficiency of olefiant gas as exhibited by the analyses may very probably be owing to the different solubility of the two gases in the sulphuretted water which was used as the confining medium during their purification from the vapour of iodide of ethyl.The action of light upon iodide of hydrogen and iodide of ethyl is therefore perfectly allalogous; in the one case we have the simple radical hydrogen eliminated and in the other the compound radical ethyl. This reaction is also perfectly parallel with that produced by the action of heat upon iodide of ethyl in presence of zinc,* except that in this last decomposition a considerably larger portion of the ethyl is transformed into hydridc of ethyl and olefiant gas. The proportion of ethyl which unde;.goes this transformation in the two reactions just mentioned and in one which I describe below is worthy of remark it is best seen from the composition of the gases left after the action of fuming sulphtiric acid.I Gases evolved by action of zinc upon iodide of ethyl C H C H H=50*03 25.79 = 1.94 1.00 11. Gases evolved by action of light upon iodide of ethyl in presence of mercury C4H C4H5H=79*lO 20.90 = 3-78 1.00 111. Gases evolved by action of light upon iodide of ethyl in presence of mercury and water C Hi C,H5.H = 11.535 3.055 = 3.78 1.00 As the hydride of ethyl occupies exactly the same volume as the ethyl from which it is derived it is evident that in the first reaction * Chem. Soe. Qu. J. 11 281. 332 DR. FRANKLAND'S RESEARCHES exactly &rd and in the second and third very nearly +th of the ethyl evolved undergoes this transformation. It was not without interest to ascertain if the action of light upon iodide of ethyl is modified in any way by the presence of water as was the case in the corresponding reaction with zinc where the whole of the ethyl was converted into hydride of ethyl by the assumption of an atom of hydrogen from the water whilst the oxygen of the latter united with zinc to form an oxyiodide.ACTION OF LIGHT UPON IODIDE OF ETHYL IN PRESENCE OF MERCURY AND WATER. Iodide of ethyl mixed with about twice its volume of distilled water was exposed as before to the direct solar rays precisely the same phenomena were observed as when the iodide alone was used although the production of gas seemed to take place more rapidly when water was present. The gases were collected purified from iodide of ethyl vapour and examined by the methods described above.A determination of the specific gravity of the gas gave the follow- ing numbers Temperature of room . . . 20.3OC. Height of barometer . '765.9"" Height of internal column of mercury . . 12.9 , Weight of flask and gas . . 33.5639 grms. Temperature in balance-case . . 21-3OC. Weight of flask filled with dry air . . 33.4492 grms. Temperature in balance-case . 21*2oc. Capacity of flask . . . 140.65 cbe. From which the specific gravity was estimated at 1.6944. The eudiometrical analysis gave the following readings In absorption eudiometer. I. Observed Difference of Corrected vol. vol. Temp. mercury Barom'.! at 00 C. and level. lmpressure. Of gasused 181.1 20.3OC.55,mm 765*9"" 128.17 1 Vol. after action of } 153,3 18.70, 3.4 , 7'64.9, 109.25 so (dry)* Vol. after admission) o,o __ -0.00 of alcohol. 333 ON THE ORGANIC RADICALS. In combustion eudiometer. 11. Observed Difference of Corrected vol. vol. Temp. mercury Barom'. at OoC. and level. Impresswe. Vol. of gas used) 96.0 19.20~. 585*7mm 7649"" 14.59 (moist) Vol. after admission 516.5 19.40 , 144.8, 7647 , 290.87 of air (moist). I-VOL after'admission } 605.2 19.70 , 649 , 764*5, 385.24 of 0 (moist). VO~.after explosion } 562.3 20.00, 103.1, 764.2, 337f23 (moist). Vol. after absorption } 503.8 18.80, 156.4 , 765.5 , 287.08 of con fdrv). 0 \ dl Vol. after admission } 689.7 21.10 , 2.6 9 762.9 , 486.75 of H (dry).VOL after <xplosion 518.1 21.60 , 153.3, 762-5, 283-27 (moist). } From analysis No. I. it follows that toe gas contains in 100 parts Gas absorbable by SO . . . 14.76 Gas unabsorbable by SO . . 85.24 -100*00 And its perfect absorption by alcohol proves the absence of hydrogen hydride of methyl and nitrogen. Analysis No. 11. shows that 14.59 vols. of the gas left intact by fuming sulphuric acid consumed 83.57 vols. oxygen and generated 50.15 vols. carbonic acid causing a contraction on explosion equal to 48.01 vols. From a simple inspection of these figures it is evident that we have the same gaseous mixture to deal with as in the previous decomposition of iodide of ethyl without the presence of water and on applying the formula given above we obtain the following values for x and y x = 11.535 y = 3.055 14-590 Hence the gas before being acted upon by fuming sulphuric acid consisted of DR FRANKLAND’S RESEABCHES 334 Ethyl .. . 67.39 Hydride of ethyl . . 17.85 Olefiant gas . ‘ . 14.76 100*00 The determination of its specific gravity given above also confirms this result as is seen from the annexed calculation C H = 67.39 x 2*00390= 135.0430 C H H = 17.85 x 1*03652= 18.5019 C H = 14.16 x 0.96742= 14*2791 167*8240 = 1*67824 100 Specific gravity as found by experiment . 1.6944 The presence of water consequently exerts no modifying influence over the decomposition of iodide of ethyl by light the products formed with and without the presence of water being identical both in composition and relative proportion.The transformation of ethyl into hydride of ethyl when its iodide is decomposed by zinc in contact with water is therefore probably owing to the high affinity of zinc for oxygen rather than that of ethyl for hydrogen although both affinities no’doubt take part in causing the resolution of water into its elements. The above decomposition of iodide of ethyl by light depending as it does directly upon the chemical rays furnishes us with the materials for the construction of an actinometer of considerable delicacy since the volumes of gas (corrected for tension of iodide of ethyl vapour &c.) evolved in equal times would give us the relative quantities of the actinic influence falling upon a given surface during these times; thus daily or even hourly readings of the instrument could be made during the time the sun is above the horizon and a register of the actinic influence in different localities be kept with as much ease as registers of the atmospheric pressure and temperature.I have not ascertained how small a quantity of light can determine the decomposition of the iodide; but very weak diffused daylight as for instance on a very cloudy or foggy day is sufficient to produce a very considerable disengagement of gas the volume of which could of course be read off at stated times with the greatest facility and without even interrupting the action of the instrument. I have also studied the action of solar light upon the iodide of ON THE ORGANIC RADICALS.335 methyl as well as upon the iodides of ethyl and methyl in contact with the various metals which has led to the discovery of an entirely new series of organo-metallic radicals possessed of very remarkable and interesting properties; the results of these researches I hope shortly to have the honour of laying before the Royal Society. The foregoing experiments form I think another link in the chain of evidence which establishes the homology of hydrogen with the radicals of the series to which ethyl and methyl belong and the simplicity of the decomposition by which the ethyl is here separated from the iodine by the action of an imponderable agent seems to some extent to disarm of their force several of the arguments lately employed with so much ingenuity by Dr.HofmannX against the formulz which I proposed for these bodies. In order to decide upon the truth or falsity of the views entertained by MM. Laurent and Gerhardt respecting these compounds according to whom their formulz ought to be doubled and the bodies themselves classed amongst the members of the marsh-gas series Dr. Hofmann under- took the examination of the products resulting from the action of heat upon valeric acid in the hope of obtaining the hitherto unknown member of the marsh-gas family represented by the formula c8 HIO which if identical with the gas evolved by the action of zinc upon iodide of ethyl would render necessary the doubling of the formula of the latter gas ; whereas if the body C8 H, were not identical with the so-called ethyl it would afford strong evidence in favour of my formula being the correct one.Unfortunately Dr. Hofmann did not succeedin obtaining this body and thus the question was left in the same condition as before. In a former memoir? I have described two separate series of carbo-hydrogens isomeric with each other the one consisting of the bodies which I consider as the alcohol radicals and the other contain- ing the members of the marsh-gas family which I regard from the mode in which they are formed as the hydrides of these radicals; thus Radicals. Marsh-gas family or hydrides. C H { Hydride of methyl (light carburetted c H,. Methyl hydrogen) } Ethyl C H , ethyl .. . C H,. H 3, Propyl C H prop91 * . C H,. H Butyl c8 H , butyl . C8 €I,. H 3 CI HI,* H &4myl C, Hll amyl &C. &C. * Chem. Soc. Qu. J. 111 121. t Ib. 111 50. 336 DR. FRANKLAND'S RESEARCHES It is obvious on inspecting the above columns that the members on the left hand are isomeric with those on the line next below then1 in the right hand column. Now two of these bodies which are represented as isomeric in the table just given have been already obtained; viz. methyl (C H,) by the electrolysis of acetic acid% and the decomposition of iodide of methyl by zinc;? and hydride of ethyl by the decomposition of cyanide of ethyl (not perfectly anhydrous) by potassium,$ and by the action of zinc upon iodide of ethyl in presence of water,§ and it is therefore only requisite to establish the identity or isomerism of these bodies in order to test the correctness of the two views which have been proposed.As both the bodies in question are gaseous at ordinary temperatures and do not change this condition under a pressure of 20 atmospheres it is evident that their physical properties cannot assist us in deciding the question; besides even assuming them to be isomeric I should not expect any difference either in their boiling points or specific gravities. Under these circumstances chlorine appeared to be the agent best suited to determine the point at issue; because although we could only expect to obtain substitution products yet the nature of these products must at once give us the key to the atomic weight of the bodies before us; for if we found +th of the hydrogen substituted by chlorine then according to the usual interpretation of this pheno- menon we must assume the simplest atom to contain 6 atoms of hydrogen and if under similar circumstances ird of the hydrogen were substituted we should have an equal right for judging the simple atom to contain only 3 equivalents of hydrogen The action of chlorine upon the so-called hydride of ethyl in diffused daylight has already been studied by Kolbe and myself,ll and we find that 1 vol.of hydride of ethyl and 1 vol. of chlorine give 1 vol. hydrochloric acid and 1 vol. of a gas having the formula C H6C1 but which is oiily isomeric and not identical with chloride of ethyl ;we proposed for it the formula which represents 1 atom of methyl conjugated with another atom of the same group in which 1 atom of hydrogen has been replaced by chlorine.From considerations deduced from the production of hydride of ethyl I now regard this chlorine compound as C cfH H > * Chem SOC.Qu. J. 11 173. $ Chem. SOC.Qu. J. I 60. t Ibid. 11 267. 5 Ibid 11 288. 11 Chem. SOC. Qu. J. I 66. ON THE ORGANIC RADICALS. hydride of ethyl in which 1 atom of hydrogen in the group C H has been replaced by chlorine; and this view explains why the body in question is isomeric and not identical with chloride of ethyl. However in whatever way this compound is viewed it is evident that &th of its hydrogen has been replaced by chlorine.It was now necessary to study the action of chlorine upon the body to which I assign the formula C H,. As there are difficulties which I have not yet been able to overcome in the way of procuring this body pure by the action of zinc upon iodide of methyl 1 employed Rolbe's method by the electrolysis of acetic acid. The apparatus used was the same as that described by that chemist in his memoir except that the gases evolved from the positive pole were allowed to stream through a long series of bulbs filled with a solution of caustic potash by which every trace of carbonic acid was removed; behind this bulbed tube were fixed threc Liebig's potash apparatus the first filled with fuming sulphuric acid the second with solution of potash and the third with concentrated sulphuric acid the last being employed to dry the gas perfectly before it passed into the tubes* which were destined afterwards to be used for the experiments the system of tubes terminated in a delivery-tube leading to the mercury- trough.As soon as the gas evolved in this last was perfectly absorbed by recently-boiled alcohol a specimen was collected for analysis and the tubes now filled with the pure gas were hermetically sealed at one end and the caouchouc connecter at the other being securely tied cut and covered with melted wax they were taken asunder and reserved for the subsequent experiments. The specimen of the gas collected as just described was exploded with excess of oxygen and gave the following numbers.Difference of Coriected vol. Observed Temp. mercury Barornr. at 00 C. and vol. level. lm pressure. Of gas used 1100.0 lfj~5OC. 576.4"" 752.5~~~ 15.28 (moist). Vol. after admission of o (moist). 1323.3 17.0°, 330.7 , 751-8, 123.77 Vol. after explosion }2664 (moist). 17*2O, 390.5, 751.6 tf 86.83 ,, To1 afterabsbrption 206.2 16~4~ 455.8 , 749.3 ,) 57.09 of CO (dry). } Vol. after admission 521.6 of H (dry). 1 17,10, 136.0 , 748.9 , 300.83 VOL after 'eiplosion 332.5 16*80, 320.6 , 749.0 , 129.73 (moist) 1 * The form of these tubes is showii in a figure given in t.his JOUIX. Vd.I p. 6 Fig. I e-f. VOL. 111.-NO. XIT. z 338 DR. FRA4NKLAND'S RESEARCHES The proportion between the voluines of combustible gas osygeir consumed and carbonic acid generated may therefore be thus stated VoI.of comb gas. 0 consumed. CO generated. 15.28 51.41 29.74 1 3.36 1.95 The numbers required theoi*etically for methyl are 1 3.5 2 A number of tubes some of the same and others of exactly double the capacity of those containing methyl were then prepared and filled with dry chlorine by displacement ;they were afterwards sealed at one extremity the caoutchouc eonnecter at the opposite end being securely tied and covered with melted wax. ACTION OF CHLORINE UPON AN EQUAL VOLUME OF METHYL. Two tubes of exactly the same capacity the one containing chlorine and the other methyl were quickly connected together by inserting their narrow necks into a strong caoutchouc connecter and securing them by silk ligatures ;they were placed in perfect darkness for 18 hours to allow the gases to become thoroughly mixed.On being afterwards exposed to diffused daylight the colour of the chlorine rapidly disappeared showing that combination ensued. After being allowed to stand in the light for several hours the tubes were hermetically scaled and their contents submitted to eudionietrical examination. On breaking off their extremities under mercury it was evident that no contraction of volume had taken place; but the dense fumes caused by allowing a few bubbles of the gas to escape into the atmosphere proved that hydrochloric acid was one of the products of the reaction and that the two gases had not simply united to form chloride of methyl.The contents of both tubes were trans- ferred into a eudiometer and the volume of hydrochloric acid was estimated by absorption first with a ball of aqueous tribasic phosphate of soda and afterwards with a bullet of fused potash which last also dried the residual gas perfectly. The following numbers were read off Difference of Corrected vol. Observed Temp. mercury Baronir. at Oo C. and vol. level. lm pressure. ON THE ORGANIC RADICALS. Thus it appears that the products of theaction of 1 vol. of chlorine on 1 vol. of methyl are 1 vol. of hydrochloric acid and 1 vol. of another gas which iiiust necessarily liave the empirical formula C H C1 expressing 4vols. This is exactly the result obtained by Kolbe and myself itz acting with an equal volume of chlorine upon the hydride of ethyl obtaiiied by the action of potassium upon cyanide of ethyl and thus up to the present point the experiments seem to prove the identity of the two bodies-the so-called methyl and the hydride of ethyl.But these experiments admit also of a differeiit interpretation; for if we assume that the chlorine acted upon only half of the methyl employed then the following equation mould cxpress the reactinn 2vols. of chlorine acting upon 1 vol. of methyl produce 1 vol. hydro- chloric acid and 1 vol. of chlormethyl C (:<) nhich last remains mixed with the excess of inethyl employed. It would be very difficult to determine directly whether the gas in questioii is a mixture 01’ a single gas; or in other worcXs whether methyl and hydride of ethyl yield identical or isomeric results when acted upon by an equal volume of chlorine; but this question can be easily dccidcd by employing an additional volume of chlorine by which if the view of the reaction just stated be Correct the cxcess of methyl will also uiidergo the pmxss of substitation and the result should be 1vol.of inethyl gas in which 1 atom of hydrogen has been replaced by elilorine (c3(:‘2)-) and 2 vols. hydrochloric acid. H Cl ACTION OP 2 VOLUMES OF CU[LORINX UPOX 1 VOL. OF METHYL. Two tubes wei-e coirnected together the oiie having a capacity exactly twice as great as the other the first being filled with dry chlorine and the last with methyl; as in the former experiment they were excluded from light for 18 hours to secure the perfect mixture of the two gases before allowing the chlorine to act.On afterwariis bringing the tubes into diffused daylight their interior became za DR. PRBYKLAND'S RESEARCHES bedewed with minute drops of an ethereal fluid which however again disappeared after the lapse of a few minutes and on opening the tubes under mercury after the action was completed no con-traction of volume was perceived to have taken place. The contents being transferred to a eudiometer the hydrochloric acid was deter- mined as before. The following numbers were obtained Difference of Correctedvol. Observed vol. Temp. mercurylevel. Barom'. at 00 C. and lm pressure. Vol. of residual gas.Vol. of H C1. 38.19 73.52 1 1.92 Hence it follows that 1 vol methyl and 2 vols. chlorine yield 2 vols. hydrochloric acid and 1 vol. of another gas which must have the formula C H C1 expressive of 2 vols. of vapour ;but this is the formula and state of condensation of the radical methyl in which 1 atom of hydrogen has been replaced by chlorine; and the action of 2 vols. of chlorine upon 1 vol. of methyl is therefore correctly ex- pressed in the equation given above. As a final proof of the correctness of this mode of interpretation and of the isomerism of methyl and hydride of ethyl it remained only to try the action of 2 vols. of chlorine upon 1vol. of the latter body. For this purpose hydride of ethyl was procured by the action of zinc upon iodide of ethyl in presence of water a process which as I have shown,* yields that gas in a state of absolute purity.One volume of this gas perfectly dried being mixed as before described with 2 vols. of chlorine and the intimate mixture then exposed to diffused daylight combination rapidly ensued and the walls of the tubes became wetted with a considerable quantity of an oleaginous fluid which did not disappear or diminish even after the lapse of several weeks during which time the tubes had been hermetically sealed. On breaking off their ends under mercury a considerable contraction was observed to have taken place the residual gas not occupying more than 2rds of the original volume. It was transferred into a eudiometer and on being treated with a ball of phosphate of * Chem.SOC. Q J. 11 291. ON THE ORGANIC RADICALS. 341 soda was so nearly absorbed that the very small volume of gas remaining after its action could not be determined. This experiment allows us to conclude that 2 vols. of chlorine with 1 vol. of hydride of ethyl yield 2vols. hydrochloric acid and a liquid which from the volumes of the gases taking part in its formation has probably the formula C H C1 and the same percentage composition as the oil of olefiant gas (C H C1 + H Cl) ;but whether the oily liquid produced in the above reaction be identical with this body I have not been able to determine as the quantity of the gaseous hydride of ethyl which would be required to form a sufficient amount of liquid to be subsequently purified for analysis would be very great.The results of these experiments on the action of chlorine upon methyl and hydride of methyl do not agree quite so nearly with theoretical calculations as I could wish owing to a slight amount of impurity contained in the methyl as indicated by the analysis of that gas the quantity of oxygen consumed being rather too far below the theoretical volume to be accounted for by possible errors of observa- tion. I have been at great pains to remove this foreign body but without success Dr. Kolbe foiind the same difficulty when he first investigated this gas and attributed the smaller quantity of oxygen consumed to the presence of a trace of oxide of methyl. As the impurity can only be present however in very minute quantity it could scarcely have any material influence upon the results of the experiments and I therefore think they allow us safely to conclude 1st.That there exist two series of hydrocarbons of the form C €Infl,the members of the one series being isomeric with those of the other and 2nd. That the formula of the gaseous hydrocarbon obtained by the electrolysis of acetic acid is C H, its atom being represented by 2 volumes of vapour; whilst the gas procured by the action of potassium upon cyanide of ethyl (not anhydrous) and by the action of zinc upon iodide of ethyl in presence of water has the formula C H, its atom being represented by 4volumes of vapour. As soon as I have succeeded in procuring pure methyl by the action of zinc upon its iodide I intend to repeat these experiments and make them more complete Although Dr.Hofniann regards the decision of the above question as sufficient to establish the correctness of one or the other of the views which have been advanced respecting these radicals on the one hand by M&l. Laurent and Gerhardt and on the other by Dr. Kolbe and myself yet I do not deem it superfluous to offer a few remarks upon the argunients which Dr. Hofmann has used with so much DE. FRANKLAND’S RESEARCHES skill against the formulz we have proposed for the bodies in question especially as several of these arguments appear at first sight very conclusive. The objections which this chemist bas made to these bodies being considered as radicals may be thus expressed 1st.The new radicals do not combine directly with the metalloids ; none of them have been found capable of reproducing a methyl- ethyl- or amyl-compound. 2nd. The voliinie of their vapours are different from that of all other known hydrocarbons. 3rd. The boiling-points of the compounds in question are in favour of their formulix being doubled. 4<th. The decomposition of the iodides of the alcohol-radicals by zinc is not perfectly analogous to that of hydriodic acid. 5th. The formuh of these bodies require to be doubled to remove the discrepancy exhibited by the boiling-points of amylene hydride of amyl and amyl. The first objection follows naturally from the circumstance that up to the time of the isolation of these bodies we were only acquainted with one basic or electro-positive radical in a separate form; viz.cacodyl which has unfortunately been loolred upon by some chemists as a type of all other organic radicals which they therefore expected to find endowed with similar powerful affinities. Such a partial view of the essential characters of a compound radical could not have been formed from a careful coniparison of the varied properties of the simple raclicals which are undoubtedly the true types of their repre- sentatives in the organic world. A slight glance at the habits and afhties ef these elementary bodies exhibits to us the most widely- diEerciit powers of combination. Conimencing with potassium and terminating with hydrogen .gold platinum iridium and nitrogen we have a series of bodies which although they all readily pass from one form of combination to another when already combined yet when once isolated exhibit as we ascend the scale an increasing reluctance to enter into union.Taking these reactions of the simple radicals then into consideration it mould be neither difficult nor visionary to predict that their organic representatives would be found possessed of as great a variety of disposition and that we should have a corresponding series of bodies commencing with cacoclyf zinc-methyl zincethyl stibethyl &c. and terminating with the radicals of the aleohol family (the perfect representatives of hydrogen) euhibiting a siinilnr decrease of combining poxwr ; and since the organic groups arc so instable in their nature and so liable to ntctaniorphosis from ON THE ORGANIC BADICALS.343 the slightest causes that we are unable to expose them without utter destruction to the powerful influences which we can bring to bear upon an elementary body it surely ought to be a matter of no great surprise if the members of the least electro-positive extremity of the series should elude all our attempts to bring them again un- injured into combiuation. If nitrogen were decomposed at a red heat by what means could we recombine that radical when once isolatcd ? I ani at present engaged in filling up the gap in the series between cacodyl and ethyl and have been lately occupied in studying the properties of an organo-metallic radical which seems to occupy a position about niidway in the series entering into direct combina- tion with several of the metalloids but n7ith a degree of affinity iinmensely less than that exhibited by cacodyl or zinc-methyl.Dr. Hofmann objects to the present forizuh of the radicals in the second place because their vapour-volumes are different from those of all other known carbo-hydrogcns. In carefully considering this objection in all its bearings upon the subject I have been quite unable to see its force or to find in this difference of vapour-volume other than a very strong proof in favour of the bodies in question being the true radicals ; for had their vapour-volunies corresponded with those of other carbo-hydrogens it would in i~y opinion have afforded striking evidence of their being no radicals at all.Here in order to seek for analogy we must again return to the siinplc radical hydrogen which presents such close relations to these organic groups and in perfect harmony with them has its atom rep~esented by 2 volcrrnes. The action of chlorinc upon all the other hydrocarbons indicates that they contain an atom of hydrogen in combination with another group aid therefore their siogle atoms like those of the hydridcs of methyl ethyl and amyl are represciited by 4 volumes of vapour and they cannot possibly be brought forrvrzrd as analogies for controlling the foriiiuh of the radicals themsel~ ryith which the members of the series C H and C €In-6are by their properties and reactions placed out of all eonncction.But the ohjection which has the greatest apparent weight arid the one to which Dr. Hofmann evidently attaches the highest importance is the 3rd; viz. “That the boiling-points of the coinpounds iii question are in favour of their forrnulz being doubled.” The apyli-cation of the beautiful and highly-intcresting law of Professor Kopp to the controlling of the forinuh of an entirely new class of bodies should be made with great caution ;for although we can by its means unerringly predict the boiling-points of the members of tlie classes of compounds upon which that law was first founded yet thc extension DR. FRANKLAND'S RESEARCHES of the list of organic bodies has proved beyond doubt that the difference of 18O or 19OC.for each addition or subtraction of the elements C H, by no means obtains when we apply it to other classes of compounds ;in fact these discrepancies might naturally be expected from a consideration of the effect which a difference in the specific and latent heat of different atoms must have upon the thermal properties of the compound; for it could not be expected that the boiling-point of water for instance should be raised through the same number of degrees by the addition of C H, as that of other bodies having a much less specific and latent heat :hence we find that the difference in the boiling-point produced by the addition of the elements C H depends entirely upon the nature of the groups to ivliich these elements are added. In the alcohols and the series of acids C H 0, the elevation in the boiling-point produced by each addition of the dements C H varies from 16" to 21OC.giving the mean number 18*5*C. and it was principally upon these series of bodies containing in addition to water two radicals-viz. ethyl methyl &c. and oxygen that Kopp's law was founded; but if we examine bodies of more simple constitution containing two radicals without water we find the increase of the boiling-point produced by each addition of the elements C H widely different from the above number. Thus at the first step we find between oxide of methyl and oxide of ethyl a difference of at least 51°C.; for according to Gay- Lussac and Dumas oxide of ethyl boils at 35*5'C. whilst oxide of methyl is incondensible at -16OC.The iodides of ethyl and methyl differ by about 30OC. and the chlorides of ethyl and amyl by 9l0C. equivalent to 3OOC. for each term of C H, whilst the difference between chloride of ethyl and chloride of methyl is at least equally great since the boiling-point of the former is ll°C. and the latter is still gaseous at -18OC. the sulphides also exhibit a difference varying from 340 to 4i'OC. for each equivalent of C H,. From these facts it is easily perceived that the difference in the boiling-points produced by the addition or subtraction of the term C H rapidly incyeases as the complexity of the compounds decreases;therefore m7e mkht reasonably expect that the radicals themselves containing as they do only one group consisting of two elements would exhibit a stiIl greater difference which is precisely what is found to be the case; amyl and valyl differ by 47°C.and ethyl and methyl will probably be found to differ to a still greater extent. Although I am averse to drawing analogies from the habits of the hydrocarbon family C H,, coiiceiving its members to have no con- nection with the groiips in dispute pet since the vnpour-uoEunae of ON THE ORGANIC RADICALS. these bodies has been brought forward as an argument in favour of doubling the formuh of the radicals I cannot refrain from referring to the boiling-points of these hydrocarbons C H, which from the fact of their formulz having already been once doubled are not likely to undergo that process again by any discrepancies in this respect.The boiling-points of two of‘ these-viz. butyrene and valerene have been determined with tolerable exactness ; the first was found by Faraday to boil at -17*8*C.; and the last by Balard at 39W. and by myself at 35OC.; but as Balard’s amylene was still mixed with small quantities of a body having a higher boiling-point and the valerene obtained from iodide of amyl contained hyciride of amyl a compound boiling at a lower temperature perhaps the mean 37OC. would be the most correct number. Here then we have two homologous bodies differing from each other by the elements C H, and approaching as near to valyl and amyl in composition as bodies belonging to a different family well could do- for valerene and butyrene differ only from amyl and valyl by con- taining one equivalent of hydrogen less-yet their boiling-points differ to the extent of 54*8OC.,which would require us to multiply their already doubled formuls by 3 to reduce them to the standard of Kopy’s law.Boiling-point. Difference. Valerene (Clo Hlo) . 370 y1 } Butyrene (C8H8) . 17*8O , 548O , It does not therefore seem that any argument in favour of doubling the formulze of the radicals can be drawn from their boiling-points but on the contrary these boiling-points taken in connection with those of their compounds give additional evidence that the formulae assigned to them are the correct ones. A few words will suffice to remove the fourth objection which is founded upon an experiment in which Dr.Hofmann failed to produce the body zinc-hydrogen (Zn H) by passing hydrochloric acid gas over metallic zinc at an elevated temperature. The forma- tion of zinc-hydrogen under these circumstances would have com-pleted the analogy between the decomposition of chloride of hydrogen and iodide of methyl; but as zinc is only very slowly acted upon by dry hydrochloric acid gas even at a high temperature,* the acid gas * Chem. SOC. Qu. J. 111 47. 346 DR. FRANKLAND’S RESEARCHES is always in excess and consequently no other result than the one obtained could be looked for any more than we could expect to preserve potassium in a stream of hot hydrochloric acid; for the series of bodies to which zinc-amyl zinc-ethyl and zinc-methyl belong increase in the energy of their affinities as their atomic weights decrease; therefore we might predict that zinc-hydrogen if such a body exist will be endowed with still more violent reactions than zinc-methyl.Now this last is instantaneously decomposed with explosion in hydrochloric acid gas ; and therefore zinc-hydrogen with perhaps still more powerful affinities could not exist for a moment in a stream of that gas or in other words could never be formed in such an atmosphere. The action of iodide of methyl upon zinc-methyl is however but very slow even at an elevated tempera- ture to which circumstance we owe the presence of this body amongst the products of the decomposition of iodide of methyl by zinc; and if we ever obtain the body zinc-hydrogen it niust be by bringing nascent hydrogen evolved from some nearly neutral body in contact with zinc As the fifth objection ‘‘ that the formulz of these bodies should be doubled to remove the discrepances in the boiling-points of valerene hydride of amyl and amyl,” is not considered by Dr.Hof-mann himself to have much weight on accouiit of the different vapour- volumes of the compounds precluding a propcr comparison I will only remark on this head that the assimilation of an atom of hydrogen with doubling of the vapour-volume and the same assimi- lation without increase of volume are widely different circumstances from which we might naturally expect very different results as regards the boiling-points of the compounds thus produced ; and accordingly when 2 vols.amyl vapour unite vith 2 vols. hydro- gen to form 4 vols. hydride of amyl vapour a depression of the boiling-point from 155O to 3Qo C. = 125O C. takes place; but this cannot be regarded as extraordinary unless it can be proved that the boiling-point of hydrogeii is not 125O C. below that of hydride of arnyl for we are entitled to assume h priori that the boiling-point of such a body would be the mean of those of its constituents. When however 4 vols. of valerene vapour unite with 2 vols. of hydrogen to form 2 vols. of amyl vapour the 4 vols. of valerene are absorbed as it were into the two volumes of hydrogen in other words the 6 volumes are condensed to 2 and the conse- quence is that the boiling-point rises 118O C I have thus endeavonred to remove seriatim the objections which have been so ably and ingeniously made by Dr.Hofinann to thc ON THE ORG.4NIC RADICALS. formuke proposed by Dr. Kolbe and myself for the groups which we conceive to represent when isolated hydrogen-and in ethyl methyl and amyl compounds the hydrogen contained in the parallel combinations of this element. That these groups do not belong to the marsh-gas family as suggested by MM. Laurent and Gerhardt is I think proved by the action of chlorine upon methyl as detailed above; whilst the production d and action of chlorine upon the hydrides of these groups clearly indicates that these hydrides form the so-called marsh-gas family hence we should gain no single advantage by doubling the present formulze of the bodies in dispute but on the contrary we should then either have to assume the existence of a third class of isomeric compounds of undefined constitution or to double our present formula for hydro-gen and represent the simplest isolable molecule of that element by H + H we should then have FI + €i = Hydride of hydrogen.C H + C H = Methyde of methyl. C €3 + C H5 = Ethide of ethyl.* &C. &C. Or adopting the notation of Messrs. Laurent and Gerhardt these formula? would be thus written H H = Hydride of hydrogen. C H C H = Methyde of methyl. C H C H = Ethide of ethyl. &C. &C. which also shows that the views advocated by these chemists on the one hand and Dr. Kolbe and myself on the other exhibit no greater difference with respect to ethyl methyl &c.than they do with regard to hydrogen. * This view of the rational constitution of these bodies was first suggested by Mr. Brodie at a recent meeting of the Society.