TRANSACTIONS OF THE SECTIONS. 55 
Notice of Mr. Fowler’s new Calculating Machine. Communicated by 
Professor Airy. 
The origin of this machine was to facilitate calculations of the pro- 
portions in which the several divisions of a poor-law district in Devon- 
shire were to be assessed. The chief peculiarity of the machine is, 
that instead of our common decimal notation of numbers, a ternary 
notation is used; the digits becoming not tenfold but threefold more 
valuable as they were placed to the left; thus, 1 and 2 expressed one 
and two as in common, but 1 0 expressed (not ten, but) three, 1 1, four, 
1 2, five; but again 2 can be expressed by 3, with | taken from it. 
Now, let 1, written thus, with a small bar above it, mean that it is 
subtractive; then 1 2 and ¢ 1 are the same in effect, both meaning five ; 
and, for a similar reason, replacing 2 by its equivalent 1 1, we have 
five written in three several ways: 1 2, or 21, or 117; the last is the 
form used. It is obvious, that by an assemblage of unit digits thus posi- 
tively or negatively written, any number may be expressed. Thus the 
number which decimally expressed is 70, becomes in the ternary system 
2121: and 2 being equal to 1 T, this may be made 10111. In the ma- 
chine, levers were contrived to bring forward the digits 1 or T, as they 
were required in the process of calculation. A full description of the 
machine, drawn up by Professor De Morgan, was presented by the 
author. 
On a Mode of solving Cubic Equations. By Mr. Wausu. 
New Logarithmic Calculations and Views. By Wm. Hovyte. 
On a new Construction of Barometer. By P. McFar.ane. 
The author proposed to substitute the weighing of a mercurial 
column, for the measure of its length, and by contrivances in accord- 
ance with this principle, to reduce or extinguish the dispreportion which 
exists between the common barometrical linear measures, and the 
variations of physical condition on which the change of these measures 
depends. 
On the four daily Fluctuations of the Barometer. By Mr. Esry. 
When the sun rises the air begins to expand by heat; this expansion 
of the air, especially of that near the surface of the earth, lifts the 
strata of air above, which will produce a reaction, causing the barome- 
ter to rise ; and the greatest rise of the barometer will take place when 
the increase of heat in the lower parts of the atmosphere is the most 
rapid, probably about 9 or 10 a.m. The barometer from that time will 
