May 12, 1887] 



NA TURE 



}i^ 



are so slight that they might almost be attributed 

 to errors of observation, or to the use of slightly- 

 erroneous corrections for the thermometers. This is 

 not the true explanation, as the next group of curves 

 illustrates. ^-! 



Strachur is near the deepest part of Upper Loch Fyne ; 

 the water which the depression contains is cut off from 

 communication with the outside by the double doors of 

 Otter and Minard with a shallow hollow between. Eight 

 sets of observations have been made, as follows : — 



April 20 June 21 Aug. 11 -Aug. 25 



0000 

 Surface 42*6 ... 49-2 ... 54-1 ... 53-5 



Bottom 41-9 ... 44-1 ... 44-2 ... 44-2 



Sept. 27 Nov. 17 Dec. 29 Feb. 4 



o o o o 



Surface 52*4 ... 46*4 ... 41-0 ... 43'0 



Bottom 44'i ... 44"2 ... 44*7 ... 45-9 



The surface range was is^^'i, with a maximum in August ; 



Fig, a. — Skate Island. 



the range of temperature on the bottom has as yet been 

 only 4°; but it is impossible, until further observations have 

 been made, to speak definitely about this. The most re- 

 markable thing apparent from the above figures is that from 

 June to December there should only have been a change 

 of half a degree Fahrenheit in bottom temperature ; but 

 an examination of the curves in Fig. 3, will bring out 

 some other curious relations. In April a uniform tempera- 

 ture of 4i°-9 was found under 10 fathoms, and this was 

 quite analogous to all the other April observations. In June 

 the surface was found greatly warmed, but at 1 5 fathoms 

 the temperature was only half a degree higher than it was 



two months before (42°* 5) : beneath that point there had 

 been considerable rise of temperature (to 44^' i), so that 

 the phenomenon was presented of a layer of cold water 

 with warmer water above and beneath. It may be men- 

 tioned in passing that but for Negretti and Zambra's 

 outflow thermometers this singular distribution could not 

 have been traced out, perhaps not even detected ; as, using 

 the deep-sea thermometers on Sixe's principle, the natural 

 induction would have been that below 15 fathoms the 

 temperature was uniform at 42°'5. In August this mini- 

 mum had almost disappeared, though a trace of it 

 remained at 35 fathoms, the point where the August curve 

 merges with that for June. By September surface cooling 

 had begun, but below 2 fathoms and down to 50 there was 



Fig. 3. — Strachur. 



a rise of temperature. At the latter depth the temperature 

 became constant to the bottom at 44 '2 as before. Novem- 

 ber and December showed the .gradual cooling of the 

 surface, and the still more gradual motion downwards of 

 the point of maximum temperature. In December the 

 bottom water had begun to warm, and in February the 

 much attenuated maximum had reached to 45 fathoms^ 

 and the remains of summer heat had fairly influenced the 

 bottom temperature. Many more very interesting relations 

 will become apparent from the study of the interlacing 

 curves of Fig. 3, which, with some modifications, are 

 applicable also to Loch Goil, a rock basin " similar and 

 similarly situated " to Loch Fyne. 



Hugh Robert Mill. 

 ( To be continued.^ 



DR. JUNKER. 



NOT since Greely told his story to the Royal Geograph- 

 ical Society has there been so crowded and enthusias- 

 tic an audience at Burlington Gardens as assembled on 

 Monday night to welcome Dr. Junker, who, during the last 

 ten years, has done so much good work for geography and 

 science in the important region between the Upper Nile, 



