187 
been subjected to this inquiry; the only series which does not 
show any trace of a maximum value is that of the uneven 
numbered group of the daily range of the temperature; but it 
is evident from the very irregular way in which the arguments 
proceed, that here large disturbing influences have come into 
play. 
The strongest evidence of the real existence of the pheno- 
menon is afforded by the daily means of barometric pressure; 
notwithstanding the large discrepancies between the amplitudes 
of the 10 formulae the quantities m and C proceed in a very 
regular way and almost exactly in accordance with the theory 
of the method: this however might be a chance phenomenon, 
but the fact, that in either of the two controlling groups the 
amplitudes show a maximum value and that in the three 
groups the arguments for p = 1 are respectively: 
| 15420! 150911” 158945) 
1. e. about the same, which ought to be the case when, as in 
casu, 2 is very small, leaves no room for ambiguity in the 
interpretation of the results. Therefore, in deducing the most 
probable value of R, either the result of the whole series alone, 
or that of the three groups may be taken into account, but 
if the latter way is followed it must be remarked that, al- 
though the amplitude A calculated from the uneven series is 
larger than that deduced from the even series, the sums m are 
not so strictly in accordance with their theoretical values in 
the uneven groups as in the even series, because the quantities 
m, for p =— 4 and + 6, ought to be very small: in consequence 
of this, a smaller weight has to he assigned to the value 25.8205 
than to the value 23.8040. 
A somewhat fainter evidence is afforded by the next element, 
the daily range of atmospheric pressure: the amplitude is so 
very small that in Appendix. Ш of Volume IX (1) no trace 
— 
| 0) I take this oecassion to redress a lapsus calami in this Appendix ; 
t was not shown by Prof Hawn that no twenty-six-day period existed, but 
that the daily range is not subjected to a periodical variation synchronous 
With the period of the sunspot cycle. 
