LIGHTNING CONDUCTORS. 221 



in a strench twelve or fifteen inches deep, and taken to a point not 

 less than one hundred feet from the building- where damp soil 

 may be reached without an excessive amount of excavation, 

 the earth connection may be said to be perfect. By taking the 

 conductor this distance from the building not only is the risk 

 of damage due to disturbance near foundations averted, but an 

 increased area of earth connection is obtained. 



14. Except in the case of very small buildings it is advis- 

 able to have at least two connections between the roof net-work 

 and the earth. In extensive buildings the writer has made a 

 practice of bringing down six or more. In these cases it is 

 occasionally necessary, from motives of economy, to join the 

 down wires together and continue one only away to the "earth " 

 There is no difficulty in making thoroughly good, soldered 

 joints in galvanised cable on terra firma, but, if funds allow, it 

 is a much better plan to have a continuous trench round the 

 building into which all the down wires are taken and then con- 

 tinued, laid up together, away to the deep " earth." 



15. As regards the comparative cost, it is interesting to note 

 that, for the same outlay not less than twenty-five times as 

 much surface can be covered by the stranded galvanised wire 

 suggested as would be by a single copper rod of one inch in 

 diameter, such as is frequently used. In fact it is perfectly safe 

 to state that, taking into consideration the increased amount 

 of labour, and additional details and accessories, required for 

 net-work system, quite twenty times better protection may be 

 obtained for the same outlay. 



16. Many interesting illustrations of the inefficacy of single 

 conductors for complete protection may be quoted from local 

 records. In one instance, which came under the direct obser- 

 vation of the writer, some damage w T as done to the part of the 

 roof of an extensive range of buildings on.^which three indepen- 

 dent conductors were fitted. Such cases afford sufficient proof 

 that single c >nductors do not effect their purpose in taking dis- 

 charges quietly away ; on the other hand there can only be ne- 

 gative evidence that buildings protected on the net-work sys- 

 tem, with many points, are not liable to be damaged. 



17. The writer trusts that these notes will be of service and 

 would conclude by making the following suggestions : — 



R. A. Soc, No. 11, 1905 



