654 DR HUGH ROBERT MILL ON THE 



and Holy Loch is similarly unaffected. No data arc given as to Lochs Long and 

 Goil, hut from ohservations made, the tidal currents in these basins appear to be 

 very slight. In Dunoon Basin the flood stream moves at about 1 knot, and the 

 same rate prevails in the Estuary, except off Gourock, where the narrowing of the 

 channel doubles the rate of flow. At the entrance to the Gareloch, the tide swirls 

 tumultuously past Row Point at 5 knots. 



Speaking generally, the ebb stream is opposite to the flood, and of equal velocity 

 and duration. Loch Fyne is an exception, the ebb being usually about half a 

 knot more rapid than the flood, and at Otter Narrows attaining even the rate of 5^ 

 knots at springs. 



At the entrance of the Sea Area, also, there is a similar difference between flood 

 and ebb. The ebb stream through Sanda Sound meets that from the south and 

 east off Deas Point, and since the Sanda stream begins two hours before high water 

 and runs to the westward for seven hours, or until one hour before low water, it meets 

 the increasing force of the Mull of Cantyre flood stream, which commences to flow 

 strongly one and a half hours before low water. The result is a frightfully tempestuous 

 sea when there is high wind. 



Paces and overfalls are formed off several of the prominent points where the 

 channel is suddenly constricted, as at Carradale in Kilbrennan Sound, or where, as 

 at Garroch Head, the flood stream splits and two branches of the ebb tide meet. 

 Taking the average rise of high tide as 85 feet for the whole Sea Area, it follows 

 that the volume of water introduced and withdrawn at each tide is 1"1524 cubic 

 sea miles. 



Rainfall and River Discharge of Clyde Sea Area. 



The Clyde Sea Area is very poorly supplied with meteorological stations, especially 

 in the region of maximum rainfall to the north-east ; but by the kindness of Dr A. 

 Buchan, of the Scottish Meteorological Society, I have been put in possession of all 

 available data, both as regards the average rainfall for the period 1866-1885, and 

 for the period over which our observations on salinity extended in 1886 and 1887. 



Tables III. and IV. give the mean monthly rainfall for the years 1866-83 at twenty- 

 four stations, grouped into the landward and seaward portions of the Area in order to 

 facilitate comparison with the salinity results. Tables V. and VI. give the actual 

 monthly rainfall of 1886 and 1887 for the same stations. Some of the values are inter- 

 polated from the consideration of results at neighbouring stations. 



[Table 



