May 19, 1887] 



NA TURE 



57 



on, the rate of cooling becomes more nearly equalised, 

 and on approaching the spring minimum the whole mass 

 of water is at one temperature, and cooling steadily 

 throughout. 



The deep inclosed basins differ from the deep open 

 basins only in degree ; but in the same direction as 

 the deep open basins differ from the Channel. On this 

 matter I do not care to speak with so much certainty ; as, 

 the conditions in the inclosed basins being much more 

 complicated, there is more probability there than else- 

 where of local and temporary disturbances being mis- 

 taken for the normal progress of events. It appears, 

 however, that summer heating takes place more slowly 

 throughout the mass, although the surface maximum is 

 earlier ; and that in the deep, comparatively still water 

 there may be at one time the conjoint effects of more 

 than one summer and winter. 



One step further in the direction of conditioning the 

 phenomena of temperature in water is to entirely cut off 

 even superficial tidal communication with the ocean ; to 

 form, in fact, a deep inland lake. Observations made by 

 Mr. Buchanan, Mr. Morrison, and myself, on Loch 



Lomond and Loch Katrine, show that there the annual 

 march of temperature is very much what might be 

 expected from the Clyde observations ; but there is the 

 great difference of the water being fresh, and having a 

 maximum density-point varying with the depth, which 

 prevents a rigid comparison being made. 



From the temperature sections, which have been de- 

 scribed, the average temperature of the whole mass of 

 water for each trip was deduced, by measuring the areas 

 occupied by each range of 2", multiplying these by their re- 

 spective mean temperatures, adding the results together, 

 and dividing by the number representing the whole area 

 of the section. In order to ascertain the temperature of 

 the surface water, that of the superficial 2 fathoms was 

 calculated in the same way. By the kindness of Mr. 

 Buchan, of the Scottish Meteorological Society, I was 

 supplied with the mean monthly air temperature (average 

 of twenty- four years) of the Clyde sea-area, and the 

 deviations from the average for each month from 

 January 1886 to February 1887. The figures are given 

 in the accompanying table, and are expressed graphically 

 by curves in Fig. 4. 



The temperature of water is not the monthly mean, as in the case of air, but that at the time when observations were made. 



This shows that the year in which our observations 

 have been made is rather an unfortunate one ; because 

 the low temperature of spring and summer, and the high 

 temperature of autumn tended to retard the heating and 

 the cooling of the water so as to produce a curve much 

 flatter than the normal one may be expected to be. The 

 maximum of air temperature occurred between July and 

 August, that of the surface water between August and 

 September, and of the whole mass of water apparently in 

 October. The air and the whole mass of water from sur- 

 face to bottom had the same mean temperature about the 

 beginning of October ; after that date the water remained 

 warmer than the air, and the whole mass of water than 

 the 2 superficial fathoms. It is specially noticeable 

 that, while during heating the surface water is far above 

 the main mass in temperature, it is only a very little 

 b^low it during cooling. 



Knowing the mass of water in the sea-area under con- 

 sideration, it is easy to convert the temperature data into 

 terms of heat ; and, using for convenience the unit of one 

 ton of sea-water raised i° Fahrenheit in temperature, the 

 following table expresses the actual changes taking place : — 



QuANTiTV OF Heat. 



Trip. 



April 13th 2tst 



June i6th 22nd 



August 4th i2th 



September.... 22ni 29th 



November....! nth j 19th 



December. ... 23rd ! srst 



February 3rd 12th 



days 



63 

 50 

 49 

 50 

 42 

 42 



+31 

 + 3'9 

 + 3-3 

 -i'4 

 -3'9 

 -27 



S c ... 

 5: a rt 



< u-c 



millions 



+ 7,000 

 + 11,600 

 + 10,000 



— 4,000 

 — 14,000 



— 9,600 



To summarise the above and give an account of heat 

 transactions, it is sufficient to say that from April to 



September there was a gain of 1,545,000 million ton- 

 degrees, corresponding to a rise in average temperature of 

 io°3 ; while from September to February there was a 

 loss of 1,200,000 million ton-degrees, corresponding to a 

 fall in average temperature of 8""o, thus leaving 343,000 

 million ton degrees of heat to be expended by April next, 

 supposing the water to return to the state in which it was 



i?86 1887 



Jan. Feb.Mar. Apr.May June July Aug. Sep. Oct. Nov.Dec. Jan. Feb. 



■BnriHi 



iiin—iiii 



■iHHi 



Fig. 4. — Clyde Sea-.\rea. Annual march of teai;)er.ilaii;. 



in April last. These numbers for heat are of course based 

 to a considerable extent on assumptions, and must only be 

 taken as part of the preliminary discussion of the exact 

 observations already recorded. 



Observations made from March 25 to April 3, since the 

 greater part of this paper was in type, show a general 



