America's sweetest honey

Full text

Sweetness isn't everything when it
comes to honey. More important than
sweetness are the subtle flavors which
awaken memories of a childhood treat, a
day in a bee yard, or scents from a long-
gone kitchen. So, we have customers who
buy our darkest, strongest honey, while
others prefer the clearest and most mild.
Tasters have described various monofloral
honey (honey from a single flower spe-
cies) as fruity, minty, sour, woody, caramel,
warm, earthy, vegetal, or some combination
of these. But people also notice whether a
honey sample is sweet or bland. So, finding
the sweetest honey in America seemed like
an interesting challenge.
We don't really know which honey is
America's sweetest, but we have some
pretty good ideas. The sweetness of honey
depends on the proportion of the differ-
ent types of sugar in it. Honey is mostly
fructose and glucose plus some sucrose,
maltose, galactose, maybe melezitose, and
small amounts of 'higher sugars'. Each sug-
ar has a different sweetness level. The sugar
Tasting honey
Apri12017
proportions vary from one type of honey
to another so sweetness also varies. This
helps explain why one honey variety may
taste bland while another is sharply sweet.
Honey that is high in fructose tastes espe-
cially sweet because fructose has a high
sweetness factor. But the other sugars have
an effect, too.
Flowers produce fructose, glucose, and
sucrose in concentrations which vary from
species to species, and from place to place.
Floral type and (to a lesser extent) geog-
raphy can influence honey sweetness. The
sugar that plants make starts out as su-
crose. Cane sugar is 100% sucrose; maple
syrup is 97% sucrose. But sucrose may
be transformed into other sugar molecules
by chemicals produced within flowers. By
the time a bee finds it, enzymes in flow-
ers split some (or nearly all) of the sucrose
molecules into fructose and glucose. Some
of the sugars are reabsorbed by the flower,
some of the glucose combines into com-
plex sugar or starch molecules, while other
chemical activities may alter the fructose/
glucose/sucrose balance. After bees collect
nectar, they add enzymes which break apart
sucrose into equal amounts of fructose and
glucose. As a beekeeper, you are justified
telling your customers that honey is 'better'
than table sugar (sucrose) because honey
has simpler molecules and doesn't toil the
body as much when converted into energy.
To figure out which honey is the sweet-
est, we need to know the chemistry of the
various types of honey - which type has the
most fructose (the sweetest natural sugar)
and the least maltose (a very bland sugar).
Honey chemistry has been studied for de-
cades. Luckily, there is a very large study
that covers all the USA and includes hun-
dreds of samples, all tested and treated in
the same systematic way. It was conducted
by the USDA years ago.
For the USDA honey study, Dr Jonathan
White asked beekeepers across the coun-
try to send samples to his lab in Beltsville,
Maryland. White's honey composition
analysis came from a veritable Who's Who·
of American Beekeepers of the '50s and
'60s. Older beekeepers will recognize many
of the producers who sent their honey for
analysis: Harry Rodenberg of Wolf Point,
Montana; Millard Coggshall of Mineola,
Florida; and, E.H. Adee of Nebraska were
three of the many participants. Others in-
cluded Dr
c.L.
Farrar who sent honey from
Wisconsin which he described as "mostly
clover with traces of dandelion, locust, and
basswood". Frank Robinson at the Univer-
sity of Florida sent three samples - orange,
COMPOSITION
OF
AMERICAN
HONEYS
Technical
BuU~
No. tU1
A~lcultural
Research
Sertlce
UNITED STATES DEPARTMENT OF
AGRl~UL~RE
\...
White's softbound book
393

grapefruit, and Japanese bamboo honey.
His 'bamboo' honey was from Japanese
Knotweed and was high moisture (19.1 %),
dark, and had a rather high maltose content
(9.8%).
All together, 309 beekeepers sent at least
one sample for analysis. For me, it was in-
teresting to see that my own father had sent
three specimens - sweetclover, goldenrod,
and aster - taken from among his 800 hives
scattered around western Pennsylvania in
1957. My father's contribution, like that of
each participant, was voluntary. His only
reward was a copy of White's softbound
book Composition of American Honeys,'
My father gave me his book. Today, it sits
among my bee books, near my writing desk.
It contains data for 504 honey samples -
color, moisture, floral source, pH, various
sugars, acids, nitrogen, and ash, as well as
producer, county and state with beekeepers'
comments. Each sample was studied for 29
different attributes. I digitized the 14,645
data entries in that report and ran the num-
bers through statistical analysis software.
It forms the backbone of the story you are
now reading.
Released in 1962, White's study of the
chemistry and composition of American
honey is over 50 years old. A few things
have changed over the years. Analysis
techniques are better. Rather than depend-
ing on beekeepers' assessments of floral
sources, pollen analysis is used today.2 But
it's still the largest publicly available study
of American honey and the report contin-
ues to be regularly referenced by modem
researchers. Principal honey qualities have
not changed - a 2006 glycemia study, for
example, presents fructose to glucose ratios
for a series of honey which are very similar
to White's values.'
After a half century, floral sources may
still have nearly identical sugar composi-
tions, but the range and type of the most
common nectar plants has certainly changed
over the years. For example, in 1960,
ranchers made hay from 10 million acres
of seeded sweet clover. Today there are
scarcely 10,000 seeded acres, though wild
sweet clover remains an important honey
plant. Buckwheat, fireweed, mangrove,
purple loosestrife and many others nectar
plants faded in value because of changing
land use. On the other hand, White does not
have a single 'sunflower', or 'canola' entry
among his 504 samples. Canola did not ex-
ist until 19744-now it produces millions
of pounds of American honey. To fill in the
gap while searching for America's sweetest
honey, I searched recent honey composition
studies and merged them with White's earli-
er study. The additional data did not change
the 'sweetest honey' result.
As we look for America's sweetest hon-
ey, we should consider why sugar (and
honey) taste sweet in the first place. The
key is a chemical reaction between sugar
molecules and taste sensors. Humans (and
bees) respond to sugar in foods, attracted by
the taste and rewarded with quick energy.
Surprisingly, some animals can't taste sugar
394
Average Moisture (percent) for USA Honey
Map 1: Moisture in USA Honey
Average pH of Honey, by State
Average Honey PH (Low=3; High=5)
PH 4.6
4.4
4.2
4.0
3.8
Map 2: pH of USA Honey
Relationship of Honey Color with Pollen (Nitrogen) and Mineral (Ash) Content
•
•
Figure 1: Color, Nitrogen, Ash
American Bee Journal

- cats, for example, lack the proper recep-
tors. Maybe it's best - a sugar rush would be
trouble for cats quietly stalking their prey.
In recognizing tastes, bees are more like us
than cats are - bees sense sweet, sour, bitter,
and salty flavors. Compared to us, they are
less sensitive to bitter flavors, but more alert
to salt. When it comes to sugar, bees notice
sucrose diluted in water at one part per thou-
sand (2.9mM)5 while humans notice about
one part per 150 (20mM)6 This means that
bees are about seven times more sensitive to
sugar in solution than people are. Bees, you
may know, have another advantage over
humans - their gustatory receptors ("taste
buds") are on their mouths, antennae, and
the base of their forelegs. They can literally
step in sugar syrup and taste it.
The usual way that the sweetness ofvari-
ous sugars is measured in people is with a
threshold taste test. Highly diluted drops
of sugar are placed on a judge's tongue
until sweetness is reported. Although most
people can sense a 1:150 sucrose solution,
for fructose it takes only one part in 250.
Fructose can be more dilute yet still noticed
because fructose is much sweeter than su-
crose. It's a subjective test, of course. Ev-
eryone perceives sweetness differently and
our sense of taste changes with age. Some
sugars (especially fructose) taste less sweet
when heated. None of this actually chang-
es our results as we search for the sweet-
est honey. We are interested in the relative
sweetness, each sugar component of honey
compared with others.
My review of eight different studies
which used dozens of participants yielded
sweetness scores that researchers generally
accept. Scientists give sucrose an arbitrary
sweetness value of 100, then they compare
other sweeteners against it. As such, fruc-
Sweetness' Values for Sugars
Fructose 163
Sucrose 100
Melezitose 90
Glucose 74
Maltose 39
Table 1: Sugar Sweetness
Average Amount of Main Sugars in Floral Honey
Fructose38%
Glucose 31%
Maltose 7%
Sucrose 1%
Table 2: Average Sugars in Honey
Acidity (pH) of Honey
pH example honey
0battery acid
1stomach acid
2
vinegar
3orange juice clover, alfalfa, orange
4
acid rain gallberry, buckwheat, honeydew
5black coffee goldenrod, blackberry
6
milk
7
pure water
Table 3: Honey Acidity
April 2017
Density Function Distribution of Sweetness Values
<l'
.~
.,
'"
o
0
Less sweet honey;
These are honeydew.
o
6
The
107108 109 110 111 112 113 114 115 116 117 118 119 120 121
Sweetness Value
Figure 2: Distribution of Sweetness
tose has a sweetness of about 163. Glu-
cose, the other main sugar in honey, only
scores 74. It's less sweet than sucrose and
half as sweet as fructose. You need to mix
a lot more glucose into water before you'll
notice it. You can see how a high-fructose
honey is sweeter than a high-glucose honey.
However, if a high-fructose honey has large
amounts of maltose (which scores only 40
on the sweetness scale), then the honey's
sweetness is diminished. All the various
sugar proportions need to be considered be-
fore we can find the sweetest honey. Table
1 shows the sweetness of the main sugars
found in honey.
In general, different floral sources of
honey have different amounts of fructose,
glucose, sucrose, maltose, and other sug-
ars. The average sugar proportions of all of
America's honey are given in Table 2. Most
honey is close to these averages, but tupe-
lo, for example, is 43.4% fructose, 25.9%
glucose, and 1.2% sucrose. That's a lot of
fructose, but tupelo honey also has a lot of
maltose, reducing tupelo'S sweetness a bit.
This becomes a jumble of math as we aver-
age each nectar source, multiply each sugar
proportion by its sweetness factor, and then
sum up the result. I did the stats and the re-
sults are interesting.
Sweetness, related to Fructose and Maltose
Sweetness
•••
Figure 3: Relationship of sweetness to amounts of fructose and maltose in
honey.
395

Black locust bloom
But before we crown the sweetest hon-
ey, we should take a quick look at some
of the other honey attributes which White
measured and reported in Composition of
American Honeys. These include pH, color,
moisture, nitrogen, and ash. For example,
the average moisture content of American
honey produced in 1956 was 16.99% while
the following year it was 17.31 %. I checked
the US Weather Service records for those
two years. In the low-moisture honey year,
the USA was in a drought, while the next
year, the country had 30% more precipita-
tion."
If we look state by state and compare
each state's rainfall with the average mois-
ture content of its honey, we find a modest
positive correlation (p=0.455). Not surpris-
ingly, states with a dry climate produce drier
honey. Nevertheless, producers in humid
states can produce good-quality stable hon-
ey with moisture below the legal 18.6% wa-
ter level. The three rainiest continental US
Sweetness of Honey Varieties
Honey"fYpe Fructose Glucose SUcrose Maltose Sweetness
Honeydew Metcalfa (Europe)
31.00 23.90 0.10
NA
107.81
Honeydew (Europe)
32.50 26.20 0.80
NA
109.01
Honeydew
31.80 26.08 0.80 8.80 112.26
Blackberry
35.70 25.69 1.18 11.06 112.40
Dandelion (European)
37.40 38.00 0.40
NA
113.94
Aster
37.73 30.43 0.75 8.53 114.16
Basswood (European)
37.50 31.90 1.20
NA
114.69
Vetch
38.29 31.71 1.57 7.46 114.74
Basswood
37.37 31.04 1.14 6.84 114.80
Buckwheat
36.30 30.33 0.61 6.92 114.92
Citrus (American)
38.67 31.84 2.22 7.18 115.04
Sourwood
39.45 25.23 0.93 11.38 115.07
Clover
37.93 3248 1.46 6.45 115.12
Goldenrod
38.09 31.98 0.52 6.53 115.20
Canda (European)
38.30 40.50 0.30
NA
115.21
Cottoo
39.49 37.01 0.78 5.02 115.26
Clover
&
Blends
38.33 3252 1.05 6.49 115.38
Alfalfa
38.00 33.48 2.42 5.97 115.52
Sweet ClOIer
39.18 34.26 1.87 5.22 115.60
Alfalfa
&
Blends
39.14 33.46 1.77 5.96 115.75
Citrus (European)
38.70 31.40 1.20
NA
115.82
Alfalfa
&
Sweet aover
39.34 33.48 1.58 5.96 115.89
Eucaly,xus (European)
39.10 33.00 1.10
NA
115.90
Euca!yj:Xus
39.81 3232 0.92 6.17 116.30
Sunfto.ver
39.20 37.40 0.30
NA
116.60
Gallberry
39.68 30.15 0.66 7.36 116.64
Sage
40.39 28.19 1.13 7.40 117.50
Tupelo
43.39 25.91 1.26 8.27 119.99
Locust
40.67 27.14 0.99 5.91 120.26
Locust (European)
4270 26.50 2.10
NA
120.64
Table 4: Average Sweetness of honey floral varieties
396
states (Louisiana, Mississippi, Alabama)
averaged a respectableI7.6% moisture; the
three driest (Nevada, Utah, Wyoming) aver-
aged 15.9%. You can see the distribution on
Map 1.
One of honey's antimicrobial qualities is
its relatively high acid content. Clover is
the most acidic common honey in the USA,
with a pH average of 3.7. At the other end of
the spectrum, goldenrod and blackberry are
less acidic, approaching an average pH of
5.0. Table 3 puts the acid levels of honeys in
perspective. You may be surprised that the
most acidic honey is produced on the alkali
soils ofthe west, while less acidic goldenrod
and blackberry nectar is most prolific when
those plants' roots are in acid soil. (Map 2) I
don't know why this odd relationship exists.
In addition to sweetness, we often judge
honey by the strength of its aroma and fla-
vor, qualities instilled in parts per billion
by esthers, flavonoids, amino acids, and
obscure chemicals. Milder honey is usu-
ally lighter in color and has fewer minerals,
pollen granules, and residual 'ash' than dark
honey. From White's study, we see that dark
honey correlates very significantly with ash
(r = 0.665) and nitrogen (a proxy for pol-
len) level (r=0.580). (Figure 1) Maltose, a
dark-tinged sugar, is also more cornmon
in darker honey (r=0.329). Although this
quantifies honey on a Pfund scale, it doesn't
tell us why floral varieties may be purplish-
black, rooster-red, golden, pale green, or
yellow. Dr. David Ball, at Cleveland State
University, says, "Curiously, the chemicals
most responsible for the color of honey re-
main largely unknown."s
Before looking at the sweetest of Amer-
ica's honey, we might consider the bland
honeys of America. These are from a mixed
bag of floral sources as well as honeydew.
The least sweet (from our study), came from
unspecified Oregon wildflowers. This hon-
ey was only 60% as sweet as the sweetest
honey. Runner-ups included samples which
beekeepers described as "toxic" (mountain
laurel) and "poisonous" (rhododendron), as
well as a few cornmon varieties. Tulip pop-
lar, manzanita, raspberry, and clover were
represented among the 20 least sweet hon-
eys. But the bland honeys and the sweetest
honeys are actually outliers. Though we can
make much of the fact that some honey is
much sweeter than others, the truth is that
nearly all honey has about the same sweet-
ness. The dry-weight average sweetness of
most American honey is about 115 (15%
sweeter than sucrose). Most honey rates
between 113 and 117 (standard deviation,
2.13). Bland and exceptionally sweet honey
are outliers, rare treats that stand apart from
the pack, as you can see in Figure 2.
When I think of sweet honey, I think of
tupelo (Nyssa aquatica). Though it ranks
near the top in this particular survey, Amer-
ica's sweetest honey is not produced from
the tupelo tree. Instead, a sample of honey
from hives owned by Claude Rose of Madi-
son, Indiana, took top honors. Mr Rose de-
scribed his honey as "almost pure" black
locust (Robinia pseudoacacia). The sample
American Bee Journal

had a very respectable 15.8% moisture with
43.3% fructose and a maltose level of nearly
zero. Second place is another black locust
sample. It was produced in Chatham, Vir-
ginia, by Bruce Anderson. For maximum
sweetness, black locust holds first and sec-
ond spots in this review. However, among
the top ten sweetest honeys, half are tupelo
from Wiwahitchka and Apalachicola, Flor-
ida. High-scoring honeys that follow these
are spring titi (Mississippi), black willow
(Louisiana), and sage (California). Other
factors may influence the sweetness of
honey, such as the interplay of the various
sugars themselves. The only truly definitive
test would require judges tasting diluted va-
rieties of various floral honey, increasing in
strength until sweetness is reported. Such
an experiment has never been done. Nor, of
course, have all the different honeys been
surveyed - it's quite possible that America's
sweetest honey is still waiting to be found.
Footnotes
1
White, Jonathan W., Composition of
American Honeys, USDA Technical Bul-
letin 1261, 1962.
2
Kaskoniene, V., "Carbohydrate com-
position and electrical conductivity of
different origin honeys from Lithuania",
LWT - Food Science and Technology 43;
pp 801-807, 2010.
3
Ischayek, JI and M. Kern, "US honeys
varying in glucose and fructose content
elicit similar glycemic indexes", Journal
of the American Dietetic Association;
106(8):1260-2, August 2006. All of the
IschayeklKern data are slightly lower
than White's data, potentially indicating
a different analytic technique.
4
Canola Council of Canada: http://www.
canolacouncil.org, 2017.
5
de Brito Sanchez, Maria Gabri-
ela, "Taste Perception in Honey Bees",
Chemical Senses, 2011.
6Dale Purves, et
al.,
editors, Neurosci-
ence, 2nd edition, Sinauer Associates,
2001
7NOAA National Centers for Environ-
mental information, Climate at a Glance:
U.S. Time Series, Precipitation, pub-
lished January 2017, retrieved on Janu-
ary 9, 2017 from http://www.ncdc.noaa.
gov/cag/
8
Long, David W., "The Chemical Com-
position of Honey", Journal of Chemical
Education, Vol 84 No. 10, pp 1643-1646,
October 2007.
Adirondack Mountain Queens
Italian queens reared and mated in toe
"North Country" from established stock
of proven wintering ability. Our reliable
"Mountain" strain can take the cold and give
you maximum honey crops year after year.
April 2017
Naturally
:.f ~
01 01 ") 01
HopGuard® II is a natural, safe, easy-to-use
biopesticide that effectivelycontrols Varroamtes
in beehives. Produced from the cones of hop
plants, HopGuard® II is a naturally-based
alternative to traditional harsh chemicals. It's
safe for bees, honey and the environment.
HopGuard® II also provides beekeepers another option to manage Varma m·le
resistance to pesticides. Rotating products is an important tactic in preventing
resistance development.
Secure the protection of your bee colony with HopGuard® II...the natural solution
to Varroa mite control.
HopGuard® II is exclusively distributed in the USA
by Mann Lake, LId:
www.mannlakeltd.com.
BetaTec/lP
hop products
efficient by nature"
Follow
us
on
11
betatec.com
Queens
Carniolan, Italian,
Buckfast, VSH
Excellent Honey Producers
Fumidil·B Fed Colonies and Nucs
Prompt Service Guaranteed
PACKAGEBEESAND NUCS
W
L BE DELIVERED TO Mi, WI and IL
INAPRILAND MAY
397