At the top I always got sound, but at the side I got no sound, although
the mercury was shaking. I then tried to see how feeble a current was
audible in the telephone. An assistant tapped the tube while I stood out
of the way, and where I could not see. I got him to tap it gentler and
gentler, and could hear the most feeble tap. A pellet of paper was next
dropped from various heights down to an inch, and each tap was perfectly
audible in the telephone. I tried many methods, and one, purely
accidentally chosen, was a piece of glass tube which I had drawn out
into a tube about 2 mm. diameter, and then nearly closed the end of
it. I have that tube here, and you will see what an ill-shapen and
ugly-looking tube it is, but it is one of the best tubes I ever got; and
finally, I found that small bits of thermometer tube, which were simply
closed at their ends with a blow-pipe, gave very good results, and I was
able to make them useful for various purposes. I then tried mounting a
tube on the end of a speaking-trumpet and speaking to the mercury, but
got no effect. In every place where I attached the glass tube itself
to a sounding-board I got a very accurate reproduction. I put one on a
piece of ferrotype plate, and that gave really the best result I ever
got. The tube was fastened with sealing-wax, and with it I got excellent
speech heard in a Bell receiver. I tried putting in a large number of
these tubes, all in quantity, on the bottom of a ferrotype plate, but
with no advantage. I have not yet tried putting them in series, one
behind the other, so as to increase the electro-motive force, but I
think that probably would be an improvement; of course it would require
many vessels of acidulated water to dip into. The most distinct
articulate speech was obtained from an ordinary ferrotype telephone
plate, secured at the edges, and one of the glass tubes you see here
attached to it.