MR SWAN ON THE CONSTITUTION OF FLAME. 25 
observations of prismatic spectra, by the late Mr Joun Apir. The telescope 
of this instrument is attached to an arm turning on the centre of a divided circle, 
so that the axis of the telescope is always in the plane of the circle. The stand 
which carries the prism also turns, independently, on the centre of the circle; so 
that although the telescope be moved through any angle to observe the rays re- 
fracted by the prism, the prism itself suffers no displacement. The prism-stand 
has screws by which the faces of the prism are made perpendicular to the plane 
of the circle. The circle was mounted on a temporary, but firm stand, so that its 
plane was vertical ; and the distance of the prism from the flame was 15} feet 
nearly. During the observations, the prism suffered no displacement, except the 
very small one occasioned by turning it round its own axis, to bring it to its 
positions of minimum deviation for the different rays of the spectrum. The ro- 
tation necessary to effect this could not exceed, from the ascertained dispersive 
power of the prism, an angle of 2°; so that the length of the path of rays pro- 
ceeding from the flame through the prism to the eye was nearly constant; and 
hence the apparent distance of the flame, and its angular breadth, was sensibly 
the same in all positions of the telescope and prism. The telescope has a good 
object-glass of 1°6 inch aperture; and, with the eye-piece used in the observations, 
magnified twenty-one times. It is furnished with a micrometer, having a fixed 
wire in the centre of the field, and a moveable cross of wires. : 
The observations were made by first carefully adjusting the telescope to focus, 
and the prism to its position of minimum deviation. By means of a screw 
connected with the stand, the circle carrying the prism and telescope was 
moved round a vertical axis until the fixed micrometer wire coincided with one 
side of the spectrum. The moveable cross was then, by means of the micrometer 
screw, made to coincide with the other side of the spectrum; and the turns and 
parts of a turn of the screw read off. From the previously ascertained deviations 
by the prism for the principal lines of FRaunHoFER, and the theodolite reading 
when the slit was viewed directly, it was easy to place the telescope so as to 
observe rays of any given refrangibility. Those chiefly observed corresponded 
with the lines B, and G of Fraunnorer. It did not seem expedient to ob- 
serve the extreme rays of the spectrum, as the faint illumination of these rays 
rendered their observation both difficult and unsatisfactory. The flames first 
examined were those of coal-gas, burned from a common jet with a single aper- 
ture. The spectrum observed was that of the luminous incandescent carbon 
conoid of the flame. The gas was supplied directly from the street pipes; but 
it was passed through one of Mitne’s regulators, in order to guard against 
the effects of varying pressure in altering the size of the flame. The flame, 
although subject to periodical fits of elongation, and slight lateral displacement, 
speedily returned to a nearly constant condition, and the observations were upon 
the whole satisfactory. They are exhibited in the following tables :— 
VOL. XXII. PART 1. G 
