60 TRANSACTION'S OF TBtE TEXAS ACADEMY OF SCIENCE. 
is best stated in his own language: “There is one result of SarashTs and 
De La Rive’s experiments which it is difficult to reconcile with theory. 
They found that the wave length of the vibration was equal to eight 
times the diameter of the resonator; theory would leave us to expect that 
the circumference of the resonator should be a half wave length, since, 
until .'the sparks pass, the current will vanish at each end of the resonator, 
as we may neglect the capacity of the knobs, and we should expect the 
wave length to be 2 rc times the diameter, instead of eight times, as found 
by Sarasin and De La Rive.” 
The object of the present paper is to call attention to certain observa- 
tions which I have made that remove the difficulty thus pointed out. But 
at the same time certain new difficulties will be presented, which I find 
it difficult to reconcile with any accepted theory. My observations con- 
firm the statement made by Sarasin and De La Rive that the nodal dis- 
tances are entirely independent of the size of the vibrator, and are pro- 
portional to the size of the resonator. But I find this distance so little in 
excess of the length of the resonator circle that it may fairly be stated, 
that the difficulty which Thomson recognized is removed; for, instead of 
being four times the diameter of the resonator it is practically -k times 
that diameter. The resonators employed by Sarasin and De La Rive 
were all small, and I suspect that they were provided with knobs. Thom- 
son, in the passage just quoted, says we may neglect the capacity of such 
knobs. But such is not the case. To show how results may be affected by 
them, attention is called to the following: 
Upon one occasion, after I had located the nodes in a certain wire, an 
assistant, provided with a second resonator of the same length as the 
one I had used, was requested to confirm my observations. To our sur- 
prise he found four nodes, where I had found only three. Exchanging 
resonators, each confirmed the observations already made by the other. 
Thinking that possibly a mistake had been made in measuring the length 
of one or the other, the wires were removed from the frame by which 
they were carried, and stretched out side by side. They were of the 
same length, and from the same, spool of wire, the only difference being 
that one terminated in knobs while the other did not. Upon removing 
the knobs, they were found to act alike. With the knobs on, the nodal 
distances were about four times the diameter of the circle, 'or 20 per 
cent more than the circumference; while without the knobs they were 
reduced to the length of the circumference. It was then decided to make 
a number of observations on wires of different lengths, with resonators of 
different sizes, both with and without knobs, to ascertain if any general 
law could be laid down containing the relative values of nodal distances 
and the circumferences of the resonators. Many unexpected difficulties 
