Electrical Signals on Land Lines. 419 



for 550 kilometres, excluding the first observations as before, is 

 0-00132 second; from Table III. it is 0-00153 second. This 

 last value ought to have coincided with the previous one. The 

 mean of the two is 0-00142 second : substituting this value for 

 a, and 550,000 for I in the above expression, we obtain as the 

 value of c 3 0*168. 



This number should have agreed with the value 0-213 deduced 

 from Table I. The discrepancy shows that it would be of little 

 use to follow out minutely the results to be obtained from every 

 series observed by M. Guillemin. 



We are not informed of the resistance in the batteries or at the 

 receiving-end ; the resistance of the lines can only be looked on 

 as the roughest approximation. 



The state of insulation of the lines is not known, and the de- 

 flections observed are not truly proportional to the currents. 

 With all these sources of error, little dependence can be placed 

 on the accuracy of the result; but M. Guillemin has indicated 

 a method by which correct results might be obtained, and has 

 shown that the phenomena of retardation observed on land lines 

 of the usual construction in France is such as would be produced 

 by an electrostatic capacity of between say 0*15 and 025 in 

 electrostatic measure. This result is equivalent to saying that 

 the capacity of a metre length of the wire is equal to that of a 

 sphere between 0'15 and 0'25 metre radius, supposed in space 

 at a distance from all conductors ; or the capacity of a foot of 

 the wire is equivalent to the capacity of a sphere between 0'15 

 and 0*25 foot radius under the same conditions. 



This result may be compared with the theoretical capacity of 

 a perfectly insulated wire supposed to be suspended in the air 

 at uniform distances above an infinite flat conducting plane, and 

 approached by no other conductor. The capacity of the wire 

 under those circumstances would be 



c=- l 



O) 4A 



where h = the height of the wire above the plane, and D its dia- 

 meter (vide article by Professor William Thomson headed " Tele- 

 graph, Electric/' Encyclopaedia Britannica, eighth edition) . Taking 

 h = 3 metres and D = 0*004 metre, the above expression 

 gives C = 0067, or less than half the smallest value obtained 

 from experiment. It is clear that no better agreement was to 

 be expected ; for the conditions in practice are far from agreeing 

 with the postulates of the theory. The above calculated value 

 will be increased by the approach of the posts to the wires, by 

 the proximity of the other conducting-wires, by the capacity of 



