330 Application of Photography to the Self-registering 
sure at the time of commencing a trace, the long index is set ver- 
tically. By varying the ratio of the lengths of the two levers we 
may enlarge the scale of the trace to any extent; for this purpcse 
there are distances marked on the short arm, corresponding to a 
scale of three, four, and five times the actual change, and the axis 
to which the arms are fixed can be moved nearer to, or further 
from the float as required. The nature of the scale must in general 
depend upon the probable extent of the barometric changes in 
during the progress of the register. Not to interfere with the 
magnetical curve, the barometer should be so placed as to form 
its trace rather on one side of the paper, thus reducing the space 
available in that direction to about one-third of its width; but by 
a particular contrivance connected with the copper case already 
described, we are enabled, without liability to exposure, to com- 
mand nearly the entire width in the other direction. This con- 
trivance consists of a set of twelve. narrow parallel ——— 
sliders, occupying in width about 14 inch, moving ver- a 
tically in a frame attached to the upper half of the cyl- iiiliil 
inder, and capable of being raised or lowered at pleas- 
ure. At the commencement of a trace, we may suppose half of- 
them to be raised; the remainder being down, the screen at the 
top of the index completely excludes all other light from the pa- 
per than the pencil passing through its own slit; but if the barom- 
eter falls beyond a certain amount, the edge of the screen will at 
length pass beyond the last slider, leaving a portion of the paper 
fully exposed ; in sucha case, one or more sliders are put down: 
if on the other hand the barometer rises to a certain amount, the 
light will at length fall on the space covered by the sliders, and 
it becomes necessary to raise some of them; in this way, the 
black bands caused by the accidental exposure of the paper, 
under extreme movements of the barometer, may be avoided. 
The cylinder, with the two. fixed lenses, the time piece, an 
the upper end of the barometer index, are all included under a 
second or external case, provided with apertures for the admission 
of the rays of light, and with a lid at the top to allow access to the 
sliding screens of the barometer. The apertures are of the same 
width as the fixed lenses, and each provided with double sliders 
by which they can be contracted at pleasure ; that appropriated to 
the magnetical trace is protected by a long rectangular tube, the 
effect of which is so complete, that “uot the slightest difference 
can be perceived on the paper whether bright daylight is freely 
admitted into the room or wholly excluded.” 
i mae. 
WD ee 
