466 
HONEY, ANALYSIS OF 
calculated from the polarizations and re¬ 
ducing sugar contents. The results are only 
true when the reducing sugar content has 
been determined by Allihns’ method. The 
calculation is as follows: 
Multiply the direct polarization at 87° 
by 1.0315 (lOOcc. of solution at 20° ex¬ 
pands to 103.15cc. at 87°). Subtract this 
figure from the direct constant polarization 
at 20°, and then divide by the factor 
2.3919. The figure thus obtained is the 
grams of levolulose in the normal weight of 
honey. Hence to find percentage, this must 
be divided by 26. 
The percentage of levulose, so found, if 
subtracted from the percentage of invert 
sugar obtained by reduction will give very 
closely the percentage of dextrose if these 
two sugars are present in nearly equal 
amounts. If these two sugars differ widely 
in percentages an error is introduced into 
the original calculation of invert sugar and 
hence in the percentage of dextrose. The 
most accurate procedure is to reduce the 
levulose to its dextrose equivalent in cop¬ 
per-reducing power by multiplying by the 
factor 0.915. This subtracted from the to¬ 
tal inducing sugars as dextrose will give 
the true percentage of dextrose. The re¬ 
sults then of this determination, viz., per 
cent of dextrose plus per cent of levulose, 
will be greater than the percentage of in¬ 
vert sugar found by reduction, but such re¬ 
sults are correct. 
UNDETERMINED. 
The sum of the percentages of water, 
sucrose, levulose, dextrose, ash, dextrin, 
acidity, subtracted from 100 gives the per¬ 
centage of undetermined matter. This con¬ 
sists of wax particles, pollen grains, albu¬ 
minoids, proteids, tannin, essential oils, 
combined acids, and a number of other 
substances. 
PROTEIN. 
Weigh out 2 grams of the honey and 
transfer to a 500cc. Kjeldahl flask; add 
10 grams of powdered potassium sulphate 
and 25ee. of C. P. sulphuric acid. Place 
the flask in an inclined position, and heat 
below the boiling-point of the acid for 
from 10 to 15 minutes, or until frothing 
has ceased (a small piece of paraffin may 
be added to prevent extreme foaming). 
This part of the operation is tedious on 
account of the sugars in the honey. Grad¬ 
ually increase the heat until boiling is ob¬ 
tained, then continue boiling until the mix¬ 
ture is colorless or nearly so, or until oxi¬ 
dation is complete. This may take over five 
hours. Cool and add 200cc. of water, then 
neutralize with sodium hydroxide solution 
(a few drops of phenolplithalein may be 
added to the liquid to determine easily 
when enough soda has been added). Con¬ 
nect immediately with a condenser and dis¬ 
till into half or tenth normal acid. Titrate 
this with tenth normal alkali, using cochi¬ 
neal as an indicator. A blank should be 
run with the reagents and the results sub¬ 
tracted from those obtained from the honey 
before calculating the percentage of nitro¬ 
gen. The per cent of nitrogen times 6.25 
gives the per cent albuminoids or protein. 
American honeys run from 0.1 to 1.0 per 
cent, and even higher. German honeys 
average about 1.08 per cent with a range 
from 0.30 to 2.42 per cent. 
DETECTION OF COMMERCIAL GLUCOSE. 
The dextro rotation of a honey at 87 
is due to honey dextrins. These are dif¬ 
ferent in character from those dextrins 
obtained by the acid hydrolysis of starch, 
or such as occur in commercial glucose. 
One point of difference is the fact that 
honey dextrins are not colored by iodin 
solution, while the dextrin of glucose, ex¬ 
cept those of high-conversion products, are 
colored by iodin. Due to this fact Beck¬ 
man has proposed the following test, which 
is qualitative in nature. 
Prepare a one-to-one solution of honey 
with water and add a few drops to 2cc. of 
iodin solution. If commercial glucose be 
present the solution turns red or violet. 
The depth and quality of the color depend 
upon the quantity and nature of the glu¬ 
cose employed for adulteration. A blank 
test with a pure honey of about the same 
color, using the same quantity of iodin solu¬ 
tion, should be made at the same time for 
the purpose of securing an accurate com¬ 
parison of color. 
If the original honey is dark in color or 
the test is not delicate enough, one can take 
the honey solution, add absolute alcohol 
until all the dextrins are precipitated, 
allow these to settle (never filter), decant 
