1905-6.] A New Form of Harmonic Synthetiser . 209 
The free end of the wire is finally Jed down to a “barograph” 
pen, which is mounted on a form of Watt’s parallel motion that 
compels it to move in a (sufficiently approximate) vertical straight 
line (Plate, fig. 2). 
In front of this arrangement an upright band of paper is made 
to slowly pass along in a horizontal direction,, and against this the 
spring carrying the “ barograph ” pen causes the latter to lightly 
press. The writing is found to be quite satisfactory, a good line 
is easily obtained, and there appears to be no appreciable stiction. 
§ 2. Description of the Mechanism which enables the Period of a 
Constituent Harmonic to be varied during motion (see 
Plate, fig. 1). 
A simple harmonic element of variable period is constructed as 
follows. A wooden truncated cone, 6 inches long, 4 inches in 
diameter at the small end and 6 at the large, has fixed in it a 
steel rod which serves as axis, and rotates between the pointed 
ends of two screws fixed in uprights from the general base. A 
similar cone is arranged with its axis parallel to that of the first, 
but disposed so that its big end is opposite to the small end of the 
other. It is well known * that unless the belt employed to 
connect the two cones be crossed, it will only have the necessary 
length when in one position on the cones. In the present case the 
belt is not crossed, but a special tension arrangement is employed 
(see fig. 1). The belt, a leather one of circular section about T 3 g- 
of an inch in diameter, encircles the two cones B and C, and also 
passes round a deeply flanged guide pulley A. A weight hung 
from the end F of the lever DEF supporting the pulley draws 
the latter continually upwards, and keeps the belt taut. When 
in motion the position of the belt on the cones can be changed by 
sliding the lever D, and therefore the attached pulley, backwards 
or forwards along the pivot rod E. If then cone B be driven 
from the electric motor, and cone C carry the small eccentric 
pulley p round which the “summation wire” passes, it is clear 
that the periodic time of C can be varied at will relative to the 
rest of the machine by moving A. In practice the arrangement 
* Proof may be found in any work on Mechanism, e.g. in that by 
S. Dunkerley, p. 23 (edition 1905). 
PROC. ROY. SOC. EDIN., YOL. XXVI. 
14 
