there is no doubt that, when intensely heated and under certain circumstances, gaseous bodies can be made to yield continuous spectra. This, however, in no way interferes with the fixity of position of the bright lines, nor can it influence the deductions derived from this fact.

Several interesting observations have been made with respect to the changes produced in the spectra of some of the metals by increase of temperature. Let me, in the first place, show you that new lines may make their

appearance in the spectra of certain elements when the temperature is raised. Thus, for instance, if we heat lithium, either the metal or its salts, in the electric arc, we obtain a splendid blue band (see Fig. 39), in addition to the red and orange rays a and 8 seen in the flame spectrum, showing that the undulations in this particular set of vibrations have become more intense. The same phenomenon is observed in the case of the strontium spectrum, where no less than four new lines (e, n, K, and λ, Fig. 39) make their appearance on increasing the temperature of the incandescent vapour of the metal. The analogy

between the production of these more highly refrangible rays and that of the overtones or harmonics of a vibrating string will occur to all.

By reducing the temperature, and therefore the intensity of the spark, only the most prominent lines of a metallic spectrum may be seen. Thus Lockyer and Frankland have shown that the magnesium (b) lines vary in length and intensity when the electrodes are separated, so that in a certain position one of the four well-known magnesium lines disappears. We shall see the application of this observation in a subsequent lecture.

The second set of facts with regard to the effect of increased heat has reference to the changes which the spectra of compound bodies undergo when the temperature is increased. This change is clearly seen in the following experiments. Let us first put a piece of fused chloride of calcium, a common lime salt, into the colourless gas flame: we observe a peculiar spectrum, which is represented roughly on this diagram, in which the red is supposed to be on the right and the blue on the left hand (Fig. 40, No. 1, and Frontispiece, No. 9). If, however,

we now pass an electric spark over some pieces of chloride of calcium, and then look at the coloured spark, we find that the spectrum thus obtained is not the same as that observed in the flame. Here you notice the difference between these two spectra: the lower drawing gives you the spark spectrum and the upper one what we may call the flame spectrum. This difference can be readily explained. It is a well-known fact that certain chemical compounds, when they are heated up above a given temperature, decompose into their constituent elements; but that, below that temperature, these compounds are capable of existing in a permanent state. When we once get the spark spectrum, we find that no alteration in the intensity of the spark can then alter the position of those lines. The position of the red lithium line never varies, although the blue line comes out. It naturally strikes every observer that these bands seen in the flame spectrum are produced by a compound of calcium (say the oxide or chloride), which remains undecomposed at the temperature of the flame. When we increase the temperature, as in the spark spectrum, we get the true spectrum of the metal. The position of these true metallic lines never alters at all, although, owing to increased intensity in the electric spark, new lines may sometimes make their appearance. Hence we can fully rely upon the spectrum test as a proof of the presence of the particular metal.

No such change in the character of the spectra is noticed in the case of those metals whose compounds are easily decomposed: thus we do not see any such phenomenon in the alkaline metals, although it is observed in the case of barium, strontium, and calcium. Another fact which bears out the truth of this explanation has

been observed by Plücker, that, in the case of bodies whose spectra change from bands to lines on increase of temperature, a recombination of the elements occurs on cooling, and the band spectrum of the compound reappears. Many other observations crowd upon us to convince us that compound substances capable of existing in the state of glowing gas yield spectra different from those of their constituent elements. Thus the spectrum of terchloride of phosphorus exhibits lines differing from those of either phosphorus or chlorine, and the chloride and iodide of copper each yields a distinct set of bands bearing no resemblance to the bright lines of the metal.

It is, here, important to learn that a distinguished spectroscopist, Professor Ångström,' does not indorse Plücker's conclusions respecting the existence of several spectra for one element, inasmuch as the spectra observed in the Geissler's tubes with low intensity are, according to Ångström, those of compound bodies, and it is only when the discharge becomes disruptive that the constant spectrum of the element appears. Ångström further states, the results of his experiments in no way bear out Plücker's conclusions, for by successively augmenting the temperature he finds that, although the intensity of the rays varies in a most complicated manner, and even new bright lines may appear, still independently of all these changes the spectrum of each substance always preserves its individual character. Thus we find ourselves in the midst of conflicting evidence, and we must endeavour to hold our minds unbiassed until the results of further research render it possible for us to come to a decision on this important subject.

1 See Appendix A. Lect. V.

The examination of the spectrum of carbon is a subject of much interest. The character of the lines which this blue flame of coal gas and air emits was first described in the year 1857 by Professor Swan. Since that time the various spectra of the carbon compounds have been care

fully examined by Dr. Attfield, Dr. W. M. Watts, and others, and it has been found that the different compounds of this element, when brought into the condition of luminous gases, either by combustion or when heated up by the electric spark, give somewhat different spectra.1

1 See Appendix E. on the Spectrum of the Bessemer Flame.

M