302 



Experience must show what number of perfectly distinct simple 

 signals can be made, and I have scarcely a doubt but that it will be 

 much more than twenty-six. Then it will be easy to invent a letter 

 code which will use these signals with the best economy for the 

 language in which the message is to be delivered. Towards this 

 object I have commenced collecting statistics showing the relative 

 frequency with which the different simple letters, and various com- 

 binations of simple letters, occur in the English language, and I must 

 soon have information enough to guide in choosing the best code for 

 a given number of simple signals. 



The investigation leading to a measurement of the electro-mag- 

 netic unit of electricity in terms of the electro-static unit, published 

 since the commencement of the present year by Kohlrausch and 

 Weber, has given all that is required to deduce from Weber's own 

 previous experiments the measurement of the electric conductivity of 

 copper wire in terms of the proper kind of unit for the telegraph 

 problem. The data required for estimating the rapidity of action in 

 a submarine wire of stated dimensions would be completed by a 

 determination of the specific inductive capacity of gutta percha, or 

 better still, a direct experiment on the electro-static capacity of a yard 

 or two cut from the cable itself. I have estimated the retardations 

 of various electric pulses, and the practicable rate of transmitting 

 messages by cables 2400 miles long, and of certain ordinary lateral 

 dimensions, on the assumption that the specific inductive capacity of 

 gutta-percha, measured as Faraday did that of sulphur, shell-lac, &c., 

 is 2, from which it probably does not differ much. These estimates 

 have been published elsewhere (Athenaeum, Oct. 1856), and I shall 

 not repeat them until I can along with them give a table of estimates 

 for cables of various dimensions, with the uncertainty as to the 

 physical property of gutta-percha either done away with by experi- 

 ment, or taken strictly into account. 



II. Plan for rapid self-recording signals by air wires and short 

 submarine cables, 



The consideration of the preceding plans has led me to think of a 

 system of working air lines, and short submarine lines, by which 

 great rapidity of utterance, considerably greater I believe than any 

 hitherto practised, may be attained. I have no doubt but that on 



