of Edinburgh, Session 1871 - 72 . 697 
the 11 thick ” or dorsal region, hut somewhat less of nitrogenous 
constituents. 4. That there is very little difference in constitution 
between the dorsal and abdominal regions of a “ Foul ” fish or 
Kelt. But, 5. That the Kelt is a much more watery fish than the 
clean salmon; and that this is slightly owing to a deficiency in 
nitrogenous ingredients, but much more to an enormous deficiency 
of oil or fat,—which is reduced to almost a sixteenth only of its 
amount in a clean-run fish. 
I am not aware of any good authority for the prevalent notion 
that a Kelt is unwholesome food. But it is plain from the foregoing- 
analysis, that the Parisian gastronome,—who, before the late 
stringent measures against river-poaching in Scotland during close¬ 
time, consumed a large proportion of Scottish Kelts,—must have 
been indebted for his enjoyment therein much more to his cook 
than to his fish. On the other hand, it is easy to see why an Api- 
cius, whose taste has been cultivated on the banks of a Scottish 
salmon-river, should wonder how any one can imagine, that the 
delicate flavour of a fish in good condition is improved by besmear¬ 
ing it with butyraceous sauces, simple or compound. 
3, On Recent Estimates of Solar Temperature. 
By James Dewar, Esq. 
(Abstract.) 
After referring to the recent discussion on the temperature of 
the sun, in which Secchi, Zollner, Yicare, Deville, and Ericsson have 
taken part, the author proceeds to group all the known methods 
of arriving at a knowledge of high temperatures under eight 
different processes. The following table gives the names of the 
physicists who have specially employed each process, together with 
the principle on which it is founded :— 
(1.) Guyton and Daniell, Prinsep, &c.—Expansion of Solids and 
Gases. 
(2.) Draper.—Refrangibility of Light. 
(3.) Clement and Desonnes, Deville.—Specific Heat. 
(4.) Becquerel, Seamens.-—Thermo-Electricity and Electric Con¬ 
ductivity. 
