252 Proceedings of Royal Society of Edinburgh. [june 6, 
was at first employed. This was afterwards modified and greatly im- 
proved in one particular. The three locking springs clasping the 
base-plate in the original instrument were removed, and their place 
taken by two similar springs, emerging through windows in an outer 
tube and clamping the bottle by pressing on the top of the collar of 
the slip-cylinder after it had closed. 
An exact copy of all the individual observations made during th 
trip of August 1886 is given in the Eeport to the Fishery Board 
presented by Dr Gibson and me. The present paper is merely in- 
tended to summarise the results, and point out some of the more 
general bearings of these observations. 
The Moray Firth . — This great bay possesses a very interesting 
configuration. The northern shore (Caithness) is rocky and steep ; 
depth increases rapidly to over 20 fathoms, and then remains as a 
broad submarine plateau, extending southward and eastward at an 
average distance of 25 fathoms beneath the surface. The western 
shore is shallow, the slope for some miles from land being slight ; 
and the same remark applies to the western half of the south coast 
(Morayshire) ; the eastern half of this coast (Aberdeenshire) is again 
rocky, with deeper water close to. A tongue-shaped depression runs 
in from the north-east along the southern portion of the firth, forming 
a deep furrow in the plateau-like sea-bottom. This has a maximum 
depth of 100 fathoms in a hole 10 miles north of Troup Head, and 
brings water over 30 fathoms deep as a very narrow trough a con- 
siderable distance west of Burghead, and close to the south shore. 
Isolated observations at various times had shown that the salinity 
of the great mass of water in the Moray Firth approached 3 '50 per 
cent, very nearly. The density (at 15° *5 6 C.) corresponding to this 
proportion of dissolved salts is 1*0260, and the density usually 
found for both bottom and surface water by Dr Gibson in 1883, and 
by me in 1886, was from 1*0257 to 1*0259. The agreement of all 
the observations taken at intervals during three years is remarkable, 
and indicates that beyond the distance of a few miles from land the 
influence of the variations of weather (rainfall particularly), from one 
season to another, on the salinity is very insignificant. Temperature 
observations naturally do not agree so closely, for one season may 
easily be a few weeks in advance of another, or behind it ; and the 
fact that the temperature is a few degrees higher or lower at any 
