April i8, 1872] 



NATURE 



483 



or on marble, or chalk, or granite, or gypsum, &c. But 

 no such effect is produced by coal gas, or defiant or 

 marsh gas. It is evidently a question of temperature, as 

 oxygen given through coal gas shows the phosphorescence 

 well. 



IV. Far exceeding in generality the effect just noticed 

 is a really magnificent blue imai^c of the flame that starts 

 up on almost every substance with Avhich the flame is 

 brought into contact. I have already drawn attention to 

 this effect in the Phil. Mag. for November 1865, and in 

 my letter of last week pointed out how the same effect 

 has more recently formed the subject of a memoir, pre- 

 sented through M. Wurtz to the Paris Academy of 

 Sciences, the author of that paper evidently being unaware 

 that the subject had already been investigated by myself. 



The appearance is as follows : When the hydrogen 

 flame is brought either vertically or sideways, say, upon a 

 white plate or a block of marble, there instantly appears a 

 deep blue and glowing impression of the exact size and 

 shape of the hollow flame. The moment the gas is extin- 

 guished, or the flame removed to the slightest distance 

 from the solid, the effect as instantly ceases. If the flame 

 be brought successively to the same spot on the solid, the 

 effect grows fainter and finally vanishes, but instantly re- 

 appears upon an adjoining portion. 



Other combustible gases, such as carbonic oxide, or 

 marsh gas, or olehant, or coal gas, do not yield this effect, 

 nor does any lamp flame, luminous or otherwise ; nor is 

 it obtained in the oxidising flame of an ordinary blow- 

 pipe ; but it is imperfectly produced in the reducing flame 

 when coal gas is used ; it is not seen when oxygen is driven 

 through coal gas, unless the latter be in excess, and it is 

 poorer and vanishes more quickly with the oxyhydrogen 

 flame than with hydrogen alone. This blue luminosity is, 

 therefore, not a question of heat, but some property de- 

 pending either on (a) the chemical nature of hydrogen, or 

 on (/3) the physical effect of its radiation. At first I 

 thought it was the latter, and that it was a new form of 

 fluorescence, so closely did it resemble those phenomena. 

 But after a week's incessant experimenting, the true cause 

 was hunted down, and found to be dependent on the 

 former effect (a), and in every case ultimately due to the 

 presence of sulphur. A chemically clean body, or a 

 freshly broken surface, did not show the blue coloration ; 

 but after e.xposure for a short time to the air of London, 

 the substance invariably yielded the blueness ; this, how- 

 ever, was not the case when the clean surface was 

 covered by a shade, or exposed to the air of the open 

 country. The combustion of coal gas and coal fires 

 yields sulphate of ammonia, a body often deposited 

 in acicular crystals in the glass tubes in a labora- 

 tory. Sulphate of ammonia is decomposed by a hydro- 

 gen flame, and when that salt is brought into contact 

 with burning hydrogen, it permanently yields the blue 

 colorescence. Hence this body is probably the main 

 source of the blueness seen whenever a hydrogen flame 

 comes into contact with glass tubes or a dirty surface. 

 This effect must repeatedly have been seen by every one 

 who has experimented on singing flames. 



When the blueness, as is so often the case, is seentinging 

 the flame itself, without contact with any body, the sulphur 

 is derived either from the vulcanised tubing, the dust of 

 which is taken up by the passing gas ; or if the hydrogen be 

 burnt from the bottle generating it, the blueness is due to 

 the decomposition of the sulphuric acid spray, as will 

 be shown further on. 



As a chemical re-agent for detecting sulphur, the deli- 

 cacy of a hydrogen flame is extraordinary. This fact was 

 estimated as follows : — Pure precipitated silica yields no 

 tlueness with the flame ; 500 grains of silica were inti- 

 rr.ately mingled with one grain of milk of sulphur. Less 

 than a yJuth of a grain of this mixture was thrown on the 

 surface of pure water or placed upon chemically clean 

 platinum foil. The water is best, but in either case the 



blue colour (absent before) now shot forth on bringing the 

 hydrogen flame down. Tried again and again with fresh 

 portions, the effect was very evident, but quickly vanished. 

 The sulphur in a similar portion of the mixture could not 

 be detected chemically by nitro-prusside of sodium. The 

 wonderful sensitiveness of the flams may be still better 

 seen in another way. Immediately after washing, the 

 fingers show no colour when brought for a moment into 

 the flame, but if a white india-rubber tube be touched 

 ever so lightly, the fingers not only show a vivid blueness, 

 but for some time any clean object touched by them, such 

 as platinum foil, shows traces of sulphur by the appearance 

 of the blue coloration with the flame. A block of melting 

 ice continually weeps itself freefrom dust, and thus presents 

 an excellent surface upon which to try the foregoing experi- 

 ment. Or a plate of platinim, after heating to redness, may 

 be written over with a stick of sulphur. If kept covered, 

 the invisible letters may long after be traced out by sweep- 

 ing the hydrogen flame over the surface of the platinum. 



Examined through a prism, the blueness derived from 

 any source shows blue and green bands, similar to the 

 spectrum of sulphur, but 1 have noticed also a red band. 

 This mode of obtaining a sulphur spectrum suggests fur- 

 ther inquiry. White marble smeared over with a bit of 

 sulphur, or with vulcanised rubber tubing, is a convenient 

 source for obtaining the effect at pleasure. 



Some sulphates and sulphides show the blueness with 

 the flame, and are evidently decomposed by the hydro- 

 gen. Thus sulphate of soda gives no blue appearance, 

 whilst sulphate of ammonia, or alum, does. 



V. Various liquids were tried in contact with the flame. 

 Sulphuric acid was very notable. Here a magniricent 

 blue effect was observed. For persistence and brilliancy 

 of the colour, this experiment leaves nothing to be desired ; 

 the spectrum is very fine. If the liquid is in a glass dish 

 when the flame is brought vertically down, the blueness 

 lights up the glass in a lovely manner.* 



VI. But the presence of sulphur is by no means the 

 only body that a hydrogen flame reveals. The least trace 

 of phosphoj-us is detected by the production of a vivid 

 green light. It is striking to notice the wonderful sub- 

 division of matter in these experiments, and how an 

 immeasurable trace of an element can evoke pronounced 

 and apparently disproportionate effects. 



Might not this ready detection of minute quantities of 

 sulphur and phosphorus be of use in the manufacture of 

 iron ; and might not hydrogen introduced into the molten 

 metal be employed for the removal of these great enemies 

 of the iron worker ^ I speak ignorantly. 



VII. Among the range of substances I have tried, tin 

 was found to yield the most conspicuous effect, after the 

 bodies named. A fine scarlet colour is almost instantly 

 produced when the hydrogen flame is brought into con- 

 tact with tin or any alloy of tin. Tin is somewhat 

 volatile, and its spectrum is rich in red rays; The tin 

 must be clean ; or the sulphur blue, which is much brighter, 

 will mask the effect. A charming experiment may be 

 made by partially scraping a soiled surface of tin ; the 

 blue and the scarlet colours mingle, and a lovely purple 

 is the result. When a trace of phosphorus is present 

 there may be obtained a green belt encircling a rich blue, 

 then a purple zone, and finally a glowing scarlet at the 

 root of the flame. These colours, it must be remembered, 

 are not imparted to the flame, but reside on the surface 

 of the body which the flame touches. And where the 

 combustion of the hydrogen is complete, as in the upper 

 part of the flame, or in the luminous stream referred to 

 (11.), these effects are not produced, they are best developed 

 at the root of the flame. 



VIII. Passing from liquids and solids, I next tried 

 gases in contact with the flame of hydrogen. Many gases 

 imparted a colour to the flame, but here the effect was 



* With all liquids, but best with mercury, a fine musical note can be ob- 

 tained by causing the jet to dip just below the surface of the liquid. 



