72 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
It was therefore foreseen that we should gain some advantage by the 
use of undamped trains if some form of electric radiator could be 
found emitting waves continuously, like the steady note of an organ 
pipe, rather than sounds like intermittent blasts on a trumpet or 
blows on a drum. There are at least three ways in which these 
undamped oscillations can be created. The first is a mechanical 
method, viz, by a high-frequency alternator. Assuming we possess 
an alternating current dynamo giving a current of a sufficiently high 
frequency, we can connect one terminal to earth and the other to a 
radiating antenna, and then on setting the machine in operation high- 
frequency undamped currents would be created in the antenna, and 
corresponding waves radiated. To secure the best results, it is 
necessary, however, to syntonise the free-time period of the antenna 
circuit and the natural frequency of the alternator. The chief diffi- 
culty, however, is to construct a machine which shall give alternating 
currents of sufficiently high frequency and voltage with sufficient 
power and current capacity. Sixteen or seventeen years ago Prof. 
Elihu Thomson and M. Tesla built dynamos giving an alternating 
current of 10 amperes at a frequency of 10,000 to 15,000, and an 
output of about 1,000 watts. Mr. Duddell exhibited to the Physical 
Society, in April, 1905, an alternator capable of a frequency of 
120,000, but its power output was not more than 0.2 watt. I have 
on the table a small alternator made by Mr. 8. G. Brown, giving an 
alternating current having a frequency of 12,000, an E. M. F. of 
20 volts, and a power of about 50 watts. Professor Fessenden has 
recently given a description of an alternator, made for him having a 
frequency of 60,000, with an output of 250 watts, running at a speed 
of 10,000 R. P. M., and giving an E. M. F. of 60 volts. Since steam 
turbines of the Laval type are now made to run at 500 revolutions 
a second, it is not difficult to construct an inductor alternator having 
a frequency of 50,000 to 100,000. Such a type of alternator has, 
however, always a large fall in terminal potential difference if called 
upon to give out current. For this reason, a type of machine with- 
out iron in the armature is to be preferred, but then it becomes more 
difficult to balance the moving parts for very high speeds. In spite 
of some attempts, the difficulties of making and driving a high- 
frequency and high-potential alternator of any considerable output, 
say 10 kilowatt size, have not yet been overcome. Even if we could 
secure a frequency of 50,000, this corresponds to a wave of 4 miles 
in length, and special antenna arrangements are necessary to radiate 
and receive such waves. Hence the alternator method of electric 
wave production will certainly not supersede the spark method, 
although in some cases it may be practicable and useful. 
