J. Trowbridge—The Earth as a Conductor of Electricity. 189 
ment of even ten-quart Bunsen cells between’wires which run 
parallel to each other a foot apart for the distance of thirty or 
forty feet. In order to detect an inductive effect under these 
conditions, a telephone of three or four units of resistance 
must be employed. The ordinary Bell Telephone has a resist- 
ance from thirty to sixty units. For still stronger reasons it is 
impossible to hear telephonic messages by induction from one 
wire to another, unless the two wires between which induction 
is produced run parallel to each and very near to each other a 
long distance. This distance generally exceeds the distance 
at which the ordinary Bell Telephone ceases to transmit articu- 
late speech. e effects which have usually been attributed 
to induction on telephone circuits are due to the earth con- 
nections and to imperfect insulation. There wonld be no 
trouble from induction if telephone wires were enclosed in a 
heard in a field an eighth of a mile from the observatory, 
where one ground of the time circuit is located. The method 
tory and not in the direct line between the observatory and the 
Boston office, the time signals were obtained by tapping the 
earth at points only fifty feet apart. Ata distance of Ave h 
dred feet directly behind the observatory, no points five hun- 
tory and the Boston office, the time signals could not be heard 
on the trial wire of six hundred feet. This was to be expected, 
since the trial wire should have its length increased as the dis- 
