of Edinburgh, Session 1883-84. 
927 
7. On the Periodic Variation of Temperature in Tidal 
Basins. By Hugh Eobert Mill, B.Sc., F.C.S. Com- 
municated by Professor Crum Brown. (Plate XIV.) 
The periodic variations of the temperature of water have been 
studied for some months at the Scottish Marine Station at Granton, 
The only tidal basin which has been considered as yet is the 
haven formed by the irruption of the sea into Granton Quarry. It 
has an area of about 7 acres ; the tidal entrance is on the west side, 
and is so situated that no water can enter until about half tide ; 
then it comes in as a very rapid stream for about three-quarters of 
an hour, when the rate falls off, and near high water it is the same 
as that of the rising tide on the shore. The ebb is gradual until 
the tide has narrowed the channel by uncovering the sandbanks 
which lie on each side of the entrance, then it is rapid for about an 
hour and a half, after which the water runs out extremely slowly, and 
does not absolutely cease until the tide begins to re-enter ; but for 
from four to five hours the water-level inside is practically unaltered. 
The depth of the quarry at low water varies from 5 to 8 fathoms in 
the parts where observations were made, and at high water the 
depth is slightly more than 1 fathom greater. The bottom shelves 
off abruptly from the bar at the entrance. 
The variation of water temperature naturally divides itself into 
two periods — aunual and diurnal. The former can only be ascer- 
tained by daily observations continued for several years, and such 
observations of both surface and bottom temperature were com- 
menced in May 1884, and are being continued. The diurnal 
variations may be investigated by continuous hourly observations 
for a number of entire days. Such series of hourly, and in some 
cases half-hourly, observations were made twice in the month of 
June for periods of thirty-six hours each, and once in the beginning 
of July for twenty-five hours, and some interesting facts were 
brought out by this means. 
It was made evident in some preliminary trials that the Miller- 
Casella thermometer was not adapted for rapid work in shallow 
water. It requires to be immersed from ten to fifteen minutes in 
3 Q 
VOL. XII. 
