OF FLA.AIES EESULTING FROM BESSEMER PROCESSES. 
481 
The Siemens-Martin Furnace. 
The investigation was made, by the kindness of Mr. F. W. Webb, in the steel 
works of the London and North-Western Railway Company, Crewe, in the first week 
of January, 1895. 
The steel melting furnace No. 3 had been charged with 15 tons of old rails at 
11 P.M. on January the 2nd. The observations were made from about 10 a.m. on 
the 3rd January. Eye observations were first made with a direct-vision spectroscope 
through one of the charging doors of the furnace ; the charge was perfectly fluid, and 
was covered with slag, through which bubbles of gas escaped. The charge was ready 
for tapping about 3 p.m. on the same day. The only line seen in the spectrum of the 
light emitted by the gas above the charge and the furnace wall beyond was the 
yellow line of sodium seen as one bright line ; the continuous spectrum was very 
strong. Photographs of the spectrum were then taken to obtain a permanent record 
and to discover how far the spectrum extended towards the ultra-violet. 
Plate 1.—A Cadett plate stained with erythrosin after McClean’s method. 
Gas 
\o 
Four ex 2 :)osures were made with the instrument jjointing as indicated in the above 
plan of the furnace, and about 3 yards away from the door. The S23ectrum extended 
to the wave-lengths here recorded. 
A. 
1. Pointed as in A, min. . . . 3800 
2. ,, ,, ,, 2 mins. . . 3540 
3. ,, ,, B, 1 min. . . . 3720 
4. ,, ,, ,, 3 mins. . . 3470 
Plate 2.— Thomas’s “ Sandall ” plate stained with erythrosin. Same direction as 
B, the slit being about 4 feet away from the door in the first, and about 3 yards in 
the second, third, and fourth photographs. 
h. m. s. 
1. Exposed from 11 27 10 to 
2. „ „ U 31 0 „ 
3. „ „ 11 36 50 „ 
4. „ „ 11 39 10 „ 
5. Solar spectj’um for reference. 
VOL. CXCVL—A. 3 Q 
h. m. s. 
11 29 10 . 
11 35 0 . 
11 38 50 . 
11 39 25 . 
A. 
3590 
3565 
3700 
