228 Prof. J. A. Fleming on Propagation of a 
long submarine cable, say the Ireland-Newfoundland (1880) 
cable. The length (2) is 1423 knots=264°2 x 10° centims. 
The resistance per knot (R)=4°161 ohms, the capacity per 
knot (C)=0°315 microfarad, the insulation resistance per 
knot (1/K.) =400-500 megohms*. 
The value of the inductance per knot (LL) is not so 
definitely known. The copper core weighs 300 lbs. per knot, 
and is equivalent in section to a round wire of 0°34 centim. 
in diameter. As a first approximation, if we calculate the 
total inductance L from the formula for the inductance of 
a straight single wire, 
erg ao. ! 2 
al\ log. G1). + - ~ 285 
we find L=11°5 henrys, from which we deduce the average 
inductance per knot (L) to be 0-008 henry. 
But the iron armour may increase this value 100 or at 
most 200 times. Hence the value of L/R per knot may be 
as low as 0°002, or perhaps as great as 0°02 or 0-4. On the 
other hand, the value of C/K per knot is 150 nearly, and 
hence the cable is very far from being distortionless. 
If we calculate a corresponding to a value of n=170, 
p=1000, and an inductance 0°008 henry per knot, then we 
have a=(°012, and the distance for drop to half initial voltage 
is 60 knots. If we attempted to cure distortion by decreasing 
the dielectric resistance per knot to 1/150 of a megohm, or 
about 6000 ohms, then we could make 2 independent of wave- 
length, but it would be increased in magnitude to a value 
0-021, and the attenuation would be worse. On the other 
hand, if we could increase the inductance per knot to 0°8 of 
a henry, the dielectric resistance being only reduced to 2/3rds 
of amegohm per knot, the distortion would be cured, and 
the value of a reduced to 6:0021. 
The distance for decay to half initial voltage would be. 
raised to 350 knots, and a great improvement effected. This is 
only an illustration of the importance of inductance on which 
Mr. Oliver Heaviside has insisted for so many years. On 
the other hand, the difficulty is to apply the theoretical 
remedy practically in the case of submarine cables, and it is 
not easy to see how it is to be done. 
Suppose, however, that we could make the resistance per knot 
* The above data are taken from Munro and Jamieson’s ‘ Pocket-book 
of Electrical Rules and Tables,’ 4th ed. p. 286. The term nawt is used 
by some writers to denote a nautical or sea mile, a length of 2029 yards, 
whilst others employ the term Anot for the above length. Strictly 
speaking a knot is the name for a speed, viz. one nautical mile per hour. 
