of the Fishery Board for Scotland. 



467 



layer to appear at the surface of the water between the Sutors when flow- 

 ing with the iiood tide into the Cromarty Firth. 



The fact that the sample collected during ebb tide at 15 fathoms from 

 the surface at the same spot shows so low a value for D seems to make 

 it probable that inside the Cromarty Firth the interchange of water had 

 not been so complete as outside. This interchange must indeed be con- 

 siderably retarded by the narrowness of the entrance to the Cromarty 

 Firth, so that a certain amount of temporary differentiation between the 

 water in the Cromarty Firth and that outside must frequently occur. 



This appears to have been the case in 1883 when, as pointed out in my 

 first report, the samples collected in or near the mouth of the Cromarty 

 Firth gave the lowest value for D. 



A purely local peculiarity already referred to remains to be noticed. 

 The depth of water in the channel between the Sutors is much greater 

 than that outside or inside of the Sutors. At one spot it even reaches 

 over 30 fathoms, while the depth outside and inside is only about 6 or 

 7 fathoms. Now, if once the cold heavy water from the North Sea 

 gets into this deep hole, it may lie there for a long time before it mixes 

 with the waters flowing to and fro over it. The waters in this hole may 

 thus have a composition corresponding to that to be found at the bottom 

 of the northern area of the North Sea, while the mass of the water in 

 the Cromarty Firth has the composition of surface Atlantic water mixed 

 of course with fresh water. This appears to have been the state of 

 matters when the 'Garland' visited the Cromarty Firth in May 1888. 

 All the samples collected by Mr F. M. Gibson in the Cromarty Firth 

 on this occasion and subsequently examined, gave values for D (Table 

 VIII.) corresponding with what I assume to be surface Atlantic water, ex- 

 cept the two bottom samples from between the Sutors at Stations IV. and 

 VI., the one being collected at low water and the other at about 2 hoursflood. 



From the data given in Table VIII. it seems almost certain that at this 

 time the mass of the water in the Moray Firth, or at least near the head of 

 the Moray Firth, was surface Atlantic water. Excluding the two samples 

 above mentioned, the value for D varies within the narrow limits of 

 1*4578 and 1*4541 in the other thirteen samples collected at different 

 localities at varying states of the tide, and having densities ( 4 S 0 ) ranging 

 from 1018*51 to 1027*52. Nothing could more clearly show how very 

 small is the difference in the chemical composition of sea water produced 

 by admixture with even a large proportion of ordinary river water. 

 It cannot indeed be otherwise, for the total salts in river water are measured 

 in miligrams per litre, while normal sea water contains over 30 grams 

 per litre. The dilution therefore of sea water with its own volume of river 

 water means only the addition of say one or two centigrams of river salts 

 to 30 grams of sea salts. 



Summary and Conclusions. 



Water in which the total halogen calculated as chlorine bears a practi- 

 cally constant ratio to the excess of the density of the water over that of 

 pure water of equal temperature was during last summer found over a 

 great area reaching from 79° north latitude down to the North Sea in 56 c 

 north latitude. Under the conditions in which I worked, the number (D) 

 expressing this ratio may be taken as slightly over 1*4710. 



On the other, hand water which, to judge from the comparatively high 

 temperature which it still retains on reaching our shores, must have come 

 from southern latitudes flows into the North Sea from the surface of the 

 Atlantic round the north of Scotland, and also through the English 

 Channel. This water gives evidence of an equally constant, though 



