1878. ] 533 [Frazer, 
one long and two short sounds. Under the conditions of the experiment 
then it may be said to be dactylic (or anapzstic). In the cut there are 
two shorts followed by one slightly longer, and this twice repeated, though 
it is extremely probable that one set (i. e. one long and two short) consti- 
tute one complete excursion of the stylus which produces the sound. E 
resembles in the figure two Indian clubs laid with the handles together. 
This is the general character whenever seen, though the size and shape of 
the component parts are subject to variation. 
In E there is in the cut but an indistinct resemblance to this apparently 
fundamental character. I have preferred, however, to have the drawing 
made without any bias by a third person rather than risk touching it to 
agree with an hypothesis. After viewing many scores of these dents and 
comparing them with the long E, I have no hesitation in saying that 
the forms are the same though much lighter. This latter fact is also at- 
tested by the manner in which the dents are run together, for this shows 
that the intensity of the sound was not great enough to cause a vibration 
which would clear the point of the stylus from the foil. The result is that 
the groove is continuous and the parts analogous to the depressions in long 
A are indicated by a widening of the groove. 
The general resemblance between I and [is clear. As the drawings 
are not made to scale (the more accurate measurements being supplied be- 
low) the greater thinness of the characters impressed upon the foil by the 
light sounds does not distinctly appear. The appended micrometric meas- 
urements will, however, give information on this head. 
O and O are like each other and unlike any of the rest, the shorter sound 
conforming to the rule above mentioned. o 
The same remarks will apply to a comparison of U and U that were 
made above inrelation to KE. A general resemblance with the short sound 
though the depressions are more tenuous is evident in OI. 
A strong corroboration of the correctness of these symbols as indicating 
the given sounds with the especial conditions employed lies in the appear- 
ance of the record for OI, which is clearly seen to be made up of ‘‘ 0” ‘“‘ee”’ 
or as pronounced O 7. 
2 
OW resembles ‘‘o”’ ‘‘u”’ strikingly. 
It will be remembered that this drawing is not absolutely but approxi- 
mately accurate. No scale or camera lucida was used. 
Dr. E. J. Nolan, who was kind enough to draw the figures on wood, 
made his drawings entirely independently from myself, nevertheless they 
agree with those made by me in every detail except their better finish. 
The narrow canals which separate or rather connect the larger depres- 
sions must be understood to result from the comparative quiescence of the 
point of the stylus at the moment when that point of the foil passed 
under it. 
They are in fact nothing but the detached parts of the canal caused by 
pressing the point of the stylus into the helical groove. They may or may 
nor appear separating the different component parts of the same sound- 
