OF NOVEMBER 9tH, 1886. 
3 
We may now turn to the cyclone itself whicli began to 
show itself distinctly at Madras on the afternoon of Monday 
the 8th. During Monday the sea became excessively rough, 
the waves at the end of the jetty measiu-ing about 17 feet 
from trough to crest, and the surf breaking magnificently 
all along the beach. The barometer, which had been falling 
slowly since 4 p.m. on the 7th, began about noon to faU. 
much faster. The wind, however, continued light but was 
gusty, blowing from the north-west. During the night the 
barometer continued to fall rapidly and the wind veered 
round to north-west by west returning to north-west at 
5 o'clock on the 9th. The maximum velocity during the 
night was 30 miles between 5 a.m. and 6 a.m. After that 
the velocity fell off to 22 miles between 8 and 9 o'clock. It 
then steadily increased again till it reached 45 miles between 
noon and 1 p.m. This was the maximum recorded in an 
hour, but since the wind came, as in all cyclones, in gusts, 
the actual wind velocity must have frequently exceeded this. 
The heaviest gusts were about 1 p.m. when the barometer also 
reached its lowest point — 29"251 inches (reduced). This is 
the lowest reading at the Observatory, but at the Christian 
College, which is about 3^ miles from the Observatory, the 
lowest reading was 29'22 inches (corrected to the Observatory 
standard). If we take into account the position of the 
centre of the cyclone, this indicates a barometric gradient 
of rather more than one-hundredth of an inch to the mile. 
Assuming for the moment that this is correct and that the 
gradient did not increase towards the centre, and taking the 
distance of the centre fiom Madras as 42 miles, the pressure 
at the centre would have been 28"8. That it was really 
considerably less than this is evident from the Armeghon 
observations, to which we will refer presently. That the 
gradient at some distance from the centre was about 0*01 
inch per mile is confirmed by comparing the readings at 
