808 J. N. Lockyer—New Method of Spectrum Observation. 
below the platinums the green line and the flutings were seen 
without the spark, that is, at the ordinary temperature of the 
flame. On introducing the sodium into the lower part of the 
flame, the green double (A5687°2 and 5681-4) and the flutings 
were not seen, either with or without the spark. 
Eeperiment No. IV.—Same arrangements as No. I, with 
metallic sodium, and with a small blowpipe instead of Bunsen. 
In this experiment the flame spectrum showed, besides the 
yellow line (D), the green double (456872 and 568174), and 
also the flutings in the green, those in the red being absent. 
s soon as the spark was passed, the green double (A 56872 
and 5681-4) became brighter, while the flutings vanished. 
In these observations the sodium was put into the flame just 
below the platinums. When put into the bottom of the flame, 
the D line was seen alone. 
Haperiment No. V.—A. glass tube $ inch in diameter was 
prepared, about six inches in length, having two platinums 
sealed into it at a distance of four inches from each other. A 
bulb was blown at each end, so that the spectrum might be 
examined with the tube end-on. A piece of sodium was put 
into the tube, and the latter exhausted with a Sprengel pump. 
An Argand burner was placed at one end of the tube, in order 
that the absorption of the vapor, as well as its radiation, might 
be observed. The metal was then very gradually heated by @ 
Bunsen flame. 
After the heating had gone on for about twenty minutes the 
absorption line of D appeared; this gradually increased In 
intensity. 
On passing the spark along the tube, the bright lines. of 
sodium appeared, the green double (456872 and 5681°4), being 
distinguishable after D had been seen for a little time alone. 
The temperature was now increased and the absorption spec 
trum again examined. The flutings in the green gradually 
made their appearance, D increasing in intensity, the gree? 
line being invisible. Afterwards the flutings in the red 
came in. ; 
ym peeing the spark the absorption spectrum, consisting of 
the red and green flutings disappeared instantaneously, and the 
green double was seen very bright; after the passage of the 
spark D dark was much increased in breadth. » 
The quantity of hydrogen given off during the change pT 
vented the passage of the spark, and the observations had to be 
discontinued. As soon as some of this had been pumped out 
the same observations were repeated with the same results. | 
Experiment No. VI—An experiment was made with lithiam 
chloride in Bunsen flame, with the same arrangement as 1? 
Experiment No. 1, 
