[ 447 ] 
through the water, from one fide of the ^ pan to the 
other, and the water appeared luminous behind the 
ftick all the way, but gave light only where it was 
difturbed. When all the water was ftirred, the 
whole became luminous, and appealed like rniiL j 
giving a conliderable degree of light to the fides of 
the pan that contained it ; and continued to do fo for 
fome time after it was at reft. The water was moft 
luminous v^hen the ftfti had been in it about 2S 
hours, but would not give any light by being ftirred, 
after it had been in it three days. 
Experiment IL 
1 put a gallon of frefti water into one pan, and a 
gallon of fea-water into another, and alfo into each 
pan a frefh herring of about three ounces* The next 
night the whole furface of the fea-water was lumi- ^ 
nous without being ftirred, but much more fo when 
put in motion 5 and the upper part of the herring, 
which lay confiderably below the furface of the wa- 
ter was very bright. The frefti water was quite 
dark, as was alfo the fifh that was in it. Theie were 
feveral very bright luminous fpots on different parts 
of the furface of the fea-water; and the whole, when 
viewed by the light of a candle, Teemed covered with 
a greafy feum. The third night, the light of the 
fea-water while at reft was very little, if at all, lels 
than before ; and when ftirred, its light was fo great, 
as to difeover the time by a watch ; and the nfli in it 
appeared as a dark fubftance. After this, its light 
was evidently decreafing, but was not quite go^ 
before the feventh night. The frefli water, and ftlh 
in 
