256 
NATURE 
[ Fan. 11, 1883 
the ordinary railway sleepers. The auxiliary rail AB and the 
corrugated steel discs DD have sufficient flexibility that two or | 
more of the latter are simultaneously depressed by an insulating | 
collecting brush or roller carried by one or by all of the car- 
riages. Depressing any of the corrugated steel discs brings the | 
stud F, which is electrically connected with the rod AB, into 
contact with the stud G electrically connected with the well- 
insulated cable. 
As only a short piece of the auxiliary rail AB is at any 
moment in connection with the main cable, the insulation cf the 
ebonite ring EE will be sufficient even in wet weather, and the | 
cast-iron box is sufficiently bigh that the flooding of the line or | 
the depositjof snow does not affect the insulation, The insula- 
Muu 
\\ 
“Fic. 2. 
tion, however, of G, which is permanently in connection with 
the main cable, must be far better. For this purpose we lead 
the gutta-percha, or india-rubber, covered wire coming from the 
main cable through the centre of a specially formed telegraph 
insulator, and cause it to adhere to the inside of the earthenware 
tube forming the stalk, Andas, in addition, the inside of each 
contact box is dry, a very perfect insulation is maintained for 
the lead coming from the main cable. Consequently as all 
leakage is eliminated except in the immediate neighbourhood of | 
the train, this system can be employed for the very longest | 
electric railways. Fig. 2 shows a modification of the contact | 
box when the insulated rail L, instead of extending all along the | 
line, is quite short and is carried by the train, and by its motion | 
presses forwards and downwards a metallic fork on the contac s 
box, thus making contact between F and G. [Other diagram, 
were explained, illustrating modifications of the contact-boxes 
in one case the well-insulated cable is carried inside the flexible 
rail, which then takes the form of a tube, shown in Fig. 3; in 
another case the cable is insulated with paraffin oil instead of 
with gutta-percha or india-rubber, shown in Fig. 4, &c.] 
The existence of there con‘act-boxes at every 20 to 50 feet 
also enables the train to graphically record its position at 
any moment on amap hanging up at the terminus, orin a signal- 
box or elsewhere, by a shadow which creeps along the map of the 
line as the train advances, stops when the train stops, and backs 
when the train backs. This is effected thus:—As the train 
Fic. 2. 
passes along, not only is the main contact between F and G auto- 
matically made, as already described, but an auxiliary contact is 
also completed by the depression of the lid of the contact-box, and 
which has the effect of putting, at each contact-box in succes- 
sion, an earth fault on an insulated thin auxiliary wire running 
by the side of the line. And just as the position of an earth 
fault can Le accurately determined by electrical testing at the end 
of the line, so we arrange that the moving position of the earth 
fault, that is the position of the train itself, is automatically 
recorded by the pointer of a galvanometer moving behind a 
screen or map, in which is cut out a slit representing by its shape 
and length the section of the line on which the train is, as shown 
in Fig. 5. In addition, then, to the small sections of 20 feet or 
more into which our auxiliary rubbed rail is electrically divided, 
there would be certain long blocked sections one mile or several 
miles in length, foreach of which onthe map a separate galvano- 
meter and pointer would be provided. [Experiments were 
shown of the system of graphically automatically recording the 
progress of a train.] 
In the preceding systems there are several contact-boxes in 
each section of the insulated rubbed rail, and several sections of 
the insulated rail in each section of the line blocked, but in the 
next system the rubbed rail is simply divided electrically into 
long sections each of as great a length as the particular system 
employed to insulate the rubbed rail will allow. In thiscase we 
arrange that the electric connection between the main cable and 
the rubbed conductor shall be automatically made by the train 
Fic. 4. 
as it enters a section, and automatically broken as the train 
leaves a section. The model before you is divided into four 
sections, each about 11 feet in length, and you see from the 
current-detectors that as the train runs either way, it puts 
current into the section just entered, and takes off current from 
the section just left. 
[Experiments were then shown of the ease with which an 
electric train could be made to back instead of going forwards, 
by reversing the connections between the revolving armatures 
and the fixed electro-magnets of the motor; also that the acci- 
dental reversal of the field magnets of the main stationary gene- 
rator, although it had the effect of reversing the main current, 
produced no change in the direction of motion of an electric 
engine, the direction of motion being solely under the control of 
the driver. ] 
