FROM SELECTED DAYS DURING THE SEVEN YEARS 1898 TO 1904. 
317 
§ 13. It will be convenient to discuss the two tables together, and it must be 
remembered that amplitude now means the ratio borne by the actual amplitude of 
the periodic term to the mean value of the element during the day. 
Comparing corresponding results from the Bureau Central and the Eiffel Tower, we 
infer that the amplitude of the 24-hour term varies but little, if at all, with moderate 
changes of height above the ground. Its value, however, at both places was much 
lower in the later series of years than in 1893. Chauveau seems disposed to 
attribute this diminution to the fact that the efflux tube had been shortened, the 
reduction in the amplitude representing the increased disturbing influence of the 
building and other surroundings. A similar phenomenon, however, is seen at Kew in 
comparing the older and more recent data, and in an enhanced degree especially in 
winter and autumn, and at Kew the efflux tube was a foot longer in the later period 
than in the earlier. The phase angles in the 24-hour term at Kew and the Bureau 
Central both vary largely with the season, but are fairly similar at corresponding 
seasons, and the values from the earlier and later series of years are in fairly close 
agreement, notwithstanding the great differences in the amplitudes. On the other 
hand, the similarity in amplitude at the Bureau Central and the Eiffel Tower is 
accompanied by a large difference in phase, at least in summer. Chauveau infers, 
from the similarity of the summer and autumn values of a x at the Eiffel Tower to the 
winter value at the Bureau Central, that the large seasonal change in a x at the 
latter station is a phenomenon confined to the lowest strata of the atmosphere. 
Experimental evidence is, however, necessary to justify the acceptance of this 
conclusion. 
The 12-hour term shows very different phenomena. The amplitude apparently 
diminishes rapidly with the height, while the phase is comparatively little affected. 
Further, the phase angle at the Bureau Central, as at Kew, varies but little with 
the season, while both amplitude and phase angle vary but little with the epoch 
from which the data are derived. The phase angles at Kew and the Bureau Central 
do not differ much, but the angles at the former station appear consistently the 
smaller. The maxima at Kew are thus somewhat later in the day than at the Bureau 
Central, the difference in time from the mean of the three seasons in the latest sets of 
data being about half an hour. 
§ 14. In consequence apparently of the rapid diminution in its amplitude with 
height Chauveau seems disposed to regard the 12-hour term as of subsidiary 
importance, and due to special conditions in the very lowest strata, if not to purely 
local causes. A somewhat similar view has been expressed by Professor F. Exner, # 
who bases his conclusions largely on observations he has made in the tropics. At 
some places, e.g., in Ceylon, he seems to have found but little trace of the ordinary 
double maximum and minimum in the diurnal inequality. Exner attributes the 
double maximum and minimum to a thin atmospheric layer near the ground, which 
* Cf. ‘Terrestrial Magnetism,’vol. 7, 1902, p. 89, 
