Grain Sampling for DON Analysis
What is DON?
Deoxnivalenol (DON), sometimes referred to as vomitoxin, is a mycotoxin
that may be produced in wheat and barley grain infected by the Fusarium head blight (FHB
or scab) fungus,
Fusarium graminearum. The occurrence of FHB in the field does not
automatically mean that DON is present in the grain, but a high level of infected kernels in
the harvested grain means DON likely will be present. The US Food and Drug
Administration (FDA) has established DON advisory levels to provide safe food and feed.
In addition, the grain and malting industry may have more rigorous standards for DON
levels in their purchased grain.
How is DON determined?
DON is measured in grain via several laboratory methods,
such as ELISA (Enzyme Linked Immunosorbent Assay), a FluoroQuant test, or gas
chromatography-electron capture (GC-EC) and gas chromatography-mass spectrometry
(GC-MS) analytical methods. Grain elevators, private labs, grain industries, and university
testing labs may provide these tests. Regardless of methods used for analysis, the key to
getting reliable and representative DON results lies with grain sampling. Sampling error
can be a significant source of variation in DON levels.
Grain sampling for DON determination begins in the field and continues at the elevator
Field harvest effects on DON results:
Producers may reduce the risk of high DON levels
in the majority of their harvested grain by careful harvest methods. Often field margins or
low levels of the field have been shown to have higher FHB severity and DON levels, in
part due to possibly denser stands along field margins (headlands area) or because of higher
prolonged humidities in low lying areas. Producers may chose to harvest these areas
separately from the majority of their field, thus keeping more diseased grain separate from
sounder grain, and reducing the risk of higher DON levels throughout the majority of
harvested grain. In addition, producers may choose higher air speeds on their field
combines, to prevent the low test weight, scabby kernels from being harvested. Postharvest
grain cleaning also is an option for removing low test weight infected kernels,
especially if gravity tables are available.
Grain sampling at elevator or mill to get representative DON
: To achieve a more
accurate DON level estimate, it is critical that the collected grain sample be representative
of an entire truckload or bin of grain. Grain and other particles, such as weed seeds,
separate based on particle size and density as they flow and settle into a truck or bin.
Smaller, denser material often may be is found the center of the truck or bin, and this is the
material that is often higher in DON content.
For sampling from an end gate grain stream, samples from
the entire width and depth of the grain stream should be collected, not just the first and last
portion of the load. A Pelican sampler or other sampling device aids in proper sample
collection and at least four samples of the entire grain stream should be collected at
intervals to represent spatial different portions of the load.
The probe is the only sampling method approved by USDAGIPSA
(United States Dept. of Agriculture’s Grain Inspection Packers & Stockyards
Administration) for stationary lots.
Multiple (5-10) probe samples are generally
recommended to obtain the best representative samples. GIPSA has specific
recommendations for probe sampling of trucklots of barley and wheat for DON. Included
in these recommendations are descriptions of sampling probes, sampling patterns, and
sample preparation, including appropriate cleaning, dividing and grinding of the samples.
Probes should not be taken from the center or outer portions of a load because these areas
do not reflect a cross section of the load. Also, GIPSA recommends a minimum 100 gram,
cleaned sample be used for testing. GIPSA’s specific guidelines can be found at the link
listed below. Other useful web-based links on grain sampling and DON also are provided.
Testing Trucklots of Barley and Wheat for Deoxynivalenol (DON)
NORTH DAKOTA STATE UNIVERSITY
DON (Vomitoxin) in Wheat -- Basic Questions and Answers
OHIO STATE UNIVERSITY
Where to Send Grain Samples for Mycotoxin Analysis
MICHIGAN STATE UNIVERSITY
How to Sample Wheat to Accurately Determine Vomitoxin Levels
US WHEAT AND BARLEY SCAB INITIATIVE (USWBSI)
Early Detection of Deoxynivalenol in Wheat
What is DON?
Deoxynivalenol (DON), commonly referred to as vomitoxin,
is a mycotoxin that may be produced in wheat and barley
grain infected by Fusarium head blight (FHB) or scab. FHB
may infect grain heads when wet weather occurs during
the flowering and grain filling stages of plant development.
The occurrence of FHB does not automatically mean that
DON is present, but a high level of scabby kernels in the
harvested grain means DON will likely be present. Levels
of DON do not necessarily correlate with levels of physical
damage in grain.
What are the critical levels
of DON for use in food and feed?
The concentrations of DON in grain are expressed as parts
per million (ppm). One ppm is equivalent to 1 pound in 1
million pounds, 1 penny in $10,000, 1 minute in two years,
or 1 wheat kernel in 80 pounds of wheat.
The U.S. Food and Drug Administration (FDA) has
established DON advisory levels to provide safe food
and feed. Unlike aflatoxin in corn, DON is not a known
carcinogen. Furthermore, grain with DON would have to
be ingested in very high amounts to pose a health risk to
humans, but it can affect flavors in foods and processing
Human food products are restricted to a 1-ppm level
established by the FDA. This level is considered safe for
human consumption. The food industry often sets standards
that are more restrictive. DON causes feed refusal
and poor weight gain in some livestock if fed above the
advisory levels. FDA advisory levels are as follows:
1 ppm Finished wheat products, such as flour, bran
and germ, that potentially may be consumed by
humans. The FDA does not set an advisory level
for raw grain intended for milling because normal
manufacturing practices and additional technology
available to millers can substantially reduce DON
levels in the finished wheat product. However,
individual millers or food industries may have
stricter requirements than 1 ppm.
10 ppm Grains and byproducts destined for ruminating
beef and feedlot cattle older than 4 months and
for poultry, providing that these ingredients don’t
exceed 50 percent of the diet
5 ppm Grains and grain byproducts destined for swine,
providing that these ingredients don’t exceed 20
percent of the diet
5 ppm Grains and grain byproducts destined for all other
animals, providing that these ingredients don’t
exceed 40 percent of the diet
How does DON impact wheat grain
quality and product performance?
FHB infection during very early kernel development can
reduce yield by decreasing kernel numbers. Slightly later
infections cause shrunken, chalky white or discolored
scabby kernels, which often are referred to as tombstones.
Kernels infected late in their development by FHB may
show no visible damage, but still have elevated levels of
DON. A Canadian study (Sinha & Savard, 1997, Can. J. of
Plant Path. 19:8-12) of DON in FHB infected wheat kernels
found an average level of 1 to 1.2 ppm in normal appearing
kernels, 2 to 5 ppm in shriveled kernels, 174 ppm in white
tombstones, and 274 ppm in pink tombstones.
Basic Questions and Answers
Farm Management Specialist
Crops Economist/Marketing Specialist
Agricultural Engineer - Post Harvest/Structures
Beef Cattle Specialist
Crops Quality Specialist
Farm and Family Resource Management Specialist
In cooperation with the N.D. Wheat Commission
North Dakota State University, Fargo, North Dakota 58105
DON (Vomitoxin) in Wheat
The majority of the DON is present in the seed coat or
bran in wheat. For the flour miller, shrunken kernels result
in a loss of milling yield because damaged kernels are
removed to improve flour quality and reduce DON content.
Enzymes found on scabby kernels can break down protein
and consequently reduce gluten strength and adversely
affect the bread and pasta making properties of the flour.
Processing and final product quality, however, are not
significantly affected by DON per se when levels are below
1 ppm. In addition, the process of milling wheat into white
flour or durum semolina typically results in the reduction of
DON by approximately 50 percent. Therefore, many grain
handlers or processors purchase grain with DON levels up
to 2 ppm without discounts. Manufacturers of whole-grain
foods will have specifications that are more rigid.
Most export markets have specifications built into their
purchase agreements to limit DON levels. The typical
standard used by the majority of world buyers is 2 ppm
maximum. In Europe, some countries have lower limits,
such as 1 ppm in the United Kingdom and 0.5 ppm in
Norway, due to their own advisory levels. Japan has set a
maximum DON level on imported wheat of 1.1 ppm. A
complex formula was developed by the Japanese Department
of Health to limit the per capita consumption of DON.
How is DON Measured?
DON is measured using several laboratory procedures.
The most common method used by the Federal Grain
Inspection Service (FGIS) and most grain handling and
processing facilities is the immunological-antibody method
called ELISA (Enzyme Linked Immunosorbent Assay)
because it is relatively fast and cheap. The gas chromatography
- electron capture (GC-EC) analytical method is
quantitative and used to calibrate ELISA test kits. It also is
the method used by the NDSU Veterinary Science Toxicology
Laboratory for measuring DON presence in grain
samples. Factors, such as sampling differences, wheat
cleaning, sample preparation and test kit standard error,
may cause differences in results between testing laboratories,
but each lab and elevator follows approved methods
that are periodically conformance tested.
Is the sampling procedure for
DON analysis important?
The reliability of testing is greatly influenced by the
sampling procedure. To achieve a more accurate DON
level estimate, it is critical that the collected grain sample
be representative of an entire truckload or bin of grain.
Grain and other particles separate based on particle
size and density as it flows into a truck or bin. Typically, the
smaller, denser material is near the center and the larger,
lighter material is near the outside of the container.
Therefore, it is expected that there will be a variation in
the concentration of affected kernels in various portions
of a truckload. In addition, since DON levels can vary
greatly between kernels of similar size and density, it is
important to take several samples from various locations
within the load.
Probe samples should not be taken from the center
or outer portions of a load because these areas do not
reflect a cross section of the load. The samples also must
represent spatially distinct areas of the load. The probe
should collect the sample from as much of the entire depth
of the truck as possible. Four to five probes per truck are
To obtain an accurate sample from an end gate grain
stream, samples from the entire width and depth of the
grain stream should be collected, not just the first and last
portion of the load. A Pelican sampler or other sampling
device aids in proper sample collection. At least four
samples of the entire grain stream should be collected at
intervals to represent spatially different portions of the load.
What are possible strategies
for using wheat with DON?
Is grain cleaning economically feasible for removing DON?
FHB blight affects the kernel in a variety of ways that
permit the scabby kernels to be removed from good quality
wheat. The kernels may be deformed enough that they
can be removed by screening. The kernels typically have a
Wheat Cleaning Examples
Wheat Sample (lb/bu) DON (ppm)
Whole Field Sample 54.2 7.0
Light Portion 55.7 5.0
Medium Portion 58.4 3.1
Heavy Portion 59.9 1.5
Screenings 41.2 26.1
Factors sample Clean-out Sample
Bushels 1,000 110 890
(lbs/bu) 59-61 57 61-63
Damage 2-5% Unchecked <0.5%
DON level 8 ppm Unchecked 0-2 ppm
Market Price $1.70 $1.00 $3.20
Value $1,700 $110 $2,848
Increase in income from cleaning this 1,000 bushels of wheat was
$858; $110 value of clean out + $2,848 value of cleaned sample -
$400 cost of cleaning - $1,700 value of unclean sample. Cleaning
cost in this example was $0.40/bushel.
lighter test weight, so the very light kernels can be removed
by airflow. If screening and aspiration do not adequately
remove the scabby grain, the wheat can be sorted by
density with a gravity table or fluidized bed separation.
The amount of scabby grain and the amount of wheat
lost during cleaning is different for each lot of grain,
so a small quantity needs to be cleaned to determine the
economics. The cost of cleaning will typically be about
40 cents per bushel. DON can occur on kernels of normal
appearance and shriveled or tombstone kernels. Therefore,
it is not always possible to reduce the DON level by
cleaning the grain.
Is respiratory protection needed while handling
All mold spores, not only those of the Fusarium fungus,
may cause allergic reactions and breathing problems if
inhaled. Appropriate personal protective gear, such as
masks designed to keep out mold spores and grain dust,
are recommended when handling grain. Generally, these
masks are either N95 rated masks, which typically have two
straps, or respirators with HEPA filters. Masks are available
at most hardware stores for about $2.50 each.
Does DON increase in storage?
After the grain dries below a moisture level of about
22 percent, fungal growth and DON production stops.
Anytime the kernel is damaged, however, the potential for
grain deterioration during storage increases. Damaged
wheat should be stored at or below 12 percent moisture
content. Studies show that the allowable storage time for
scabby wheat is slightly less than for non-affected wheat.
Therefore, recommended airflow rates to dry scabby wheat
should be slightly increased above the rates recommended
for good quality wheat. Affected wheat should be cooled
by aeration soon after being placed in storage and further
cooled periodically as outdoor temperatures decline until
the wheat is about 25 F.
What is the best strategy for marketing wheat with DON?
The most cost effective way to market wheat with high
DON levels depends on many factors, such as how much
DON is in the wheat and the current discounts; ability
and cost to segregate, clean and/or blend the wheat; cost
of storage; contract obligations with the elevators; the loan
deficiency payment (LDP); and the price outlook.
Elevator discounts generally are the most severe at
harvest. Discounts usually decrease after harvest, as the
marketing system is able to assimilate the lower quality
grain over time. Market reaction to DON levels can vary
depending on which market elevators are selling into,
how much of their local draw area was affected and the
availability of blending stocks. Elevators and grain exporters
risk outright rejection of shipments that exceed contract
At harvest, every attempt should be made to segregate
and store on-farm DON-affected and nonaffected wheat.
The possibility of deferring contracted wheat in the hope
that discounts will lessen with time should be explored. Onfarm
storage also gives the producer time to improve quality
before delivery by cleaning and/or blending. To be worthwhile,
the combined value of the cleanout and clean grain,
less cleaning costs, must be greater than the value of the
original grain. The value of cleanout varies at elevators. A
one-time in/out charge of about 8.5 cents per bushel is an
additional storage cost that also should be considered.
A CCC loan on wheat being stored may offer benefits.
Most importantly, it provides cash flow. It also may have
a lower interest charge than a commercial loan. However,
any LDP that might be available cannot be taken if a CCC
loan is taken. Forfeiture on the loan usually is not the best
alternative since the discounts on forfeited wheat would
be 50 cents for 2.1 to 3 ppm, 75 cents for 3.1 to 4 ppm and
$1 for 4.1 to 5 ppm. Wheat more than 5 ppm would have a
settlement value of zero.
For wheat in the bin, the cost of good quality on-farm
storage is about 1.5 to 2.1 cents per month, based on
interest rates of 4.5 to 7 percent, respectively. Storage into
April/May could be profitable if distant futures prices are
at least 12 to 15 cents higher and remain there, the basis
does not deteriorate, and DON discounts are reduced or
quality can be improved.
Can DON contaminated grain be used as a livestock feed?
In most cases, wheat containing DON can be used as a
livestock feed. However, there are some classes of livestock
that do not tolerate DON well.
FDA limits for cattle are similar to poultry
(dietary levels of 5 ppm). Research conducted in North
Dakota and Minnesota has suggested growing and finishing
cattle can tolerate higher levels (up to 18 ppm based on
research at the Carrington Research Extension Center).
FDA limits the level to 2 ppm DON in the
diets of lactating dairy cows.
Do not feed wheat containing DON to gestating
or lactating sows or pigs weighing less than 50 pounds.
Growing and finishing pigs may be fed grains containing
DON, provided the level of DON in the diet does not exceed
Poultry can be fed grain containing DON. Total
dietary levels should not exceed 5 ppm (e.g., 10 ppm DON
in wheat could be fed at 50 percent of the diet).
No research data exists that has evaluated
feeding grain containing DON to horses. Since monogastric
animals tend to be more sensitive than ruminants to these
type of toxins, horse owners should be extremely cautious
about feeding DON containing grain to horses.
Recommendations for feeding wheat
In addition to the problems related to DON, feed wheat is
a grain that requires diligent feeding management to be
successfully included in a ration for cattle. Wheat ferments
very rapidly, making it a difficult to feed at high levels. Here
are a few pointers for making the most of this grain:
• Wheat is higher in protein and similar in energy to
corn (see following table).
• Limit wheat to 40 percent or less of the ration in
backgrounding and finishing diets.
• Limit durum to 30 percent or less of the ration in
backgrounding and finishing diets.
• Gradually adapt cattle to wheat-based diets. Start
with low levels (10 to 15 percent) and then gradually
increasing the wheat level up to 30 percent (durum)
or 40 percent (hard wheats).
• Wheat should be coarsely rolled or cracked, but not
finely ground for optimum performance.
• Wheat should not be fed in self feeders.
For more detailed information on grain feeding, check
out the following web sites:
Nutrient content of various feed grains
Wheat Barley Corn Oats Sorghum
TDN (%) 88 88 90 77 82
NEm (Mcal/kg) 2.18 2.06 2.24 1.85 2.00
NEg (Mcal/kg) 1.50 1.40 1.55 1.22 1.35
CP (%) 14.2 13.2 9.8 13.6 12.6
Nutrient analysis can vary. A laboratory analysis is recommended.
What can I do to prevent DON
in the future?
The environment plays a critical role in the development of
FHB and the production of DON. Humid and warm conditions
during flowering favor FHB and DON production. When
environmental conditions are ideal, multiple control practices
are needed to control this pernicious disease.
will not eliminate Fusarium head blight and
DON accumulation, but will help reduce the severity, even in
epidemic years. Research studies in other states show that
infections have been 5 to10 times higher when corn was
the previous crop than when wheat was the previous crop.
In 2005, a research study in Fargo showed that the field
severity of FHB was 2 times higher with wheat planted into
wheat stubble than when the same variety was planted the
same day in an adjacent field of soybean stubble.
No wheat varieties are totally
resistant to Fusarium head blight, but some varieties show
more tolerance to the disease and DON accumulation.
Of varieties currently available, Glenn, Alsen and Freyr
are among the most tolerant. For additional details go to
Planting at least two or three tolerant wheat varieties with
differing maturity dates will help minimize the risk of scab
infection. In a wet year, large quantities of the fungus that
causes scab are produced, subjecting virtually any wheat
field to some degree of Fusarium infection. Check variety
trial results from experimental plots to see which varieties
perform best in the presence of the disease.
can help reduce Fusarium head blight and
DON levels by 50 percent to 70 percent in North Dakota
in most years, when using the best available fungicides
and appropriate application timings. Success is greatest
when fungicides are applied to moderately susceptible to
moderately resistant spring wheat varieties. Under severe
epidemics, fungicides have not sufficiently reduced disease
or DON levels to achieve a top market grade in barley or in
very susceptible wheat and durum cultivars. For additional
information on the use of fungicides to control FHB, go to
Seed treatments before planting may improve seed
germination and seed vigor, but they will not prevent FHB
infection or DON accumulation.
buries disease-carrying debris, allowing for
microbial degradation of the Fusarium fungus and lowering
the chance of fungal spore dispersal. Moldboard plowing
is more successful at burying residue than chisel plowing.
However, other factors such as soil or water erosion are
important considerations for tillage practices and residue
concerns may be more effectively dealt with by crop rotation.http://scabusa.org/pdfs/forum_01_proc_fstu.pdf
Deoxynivalenol (DON) is a toxin produced
by fusarium fungi. DON occurs in barley and
other feed grains when grown under certain
climatic conditions. The quality of beer can
be adversely affected when malting barley with
significant concentrations of DON are used.
Illnesses have been observed in livestock that
have consumed feed grains containing high
levels of DON concentrations. As a consequence,
wheat and barley lots often receive
price discounts when DON concentrations
exceed certain levels.
Levels of DON are typically measured in the
marketing channels with commercially available
test kits. Test kits typically are based on
enzyme linked immunosorbent assay (ELISA)
technology. DON measurements are often
made on each lot delivered to the market by
producers. Due to the significant economic
consequences, the accuracy and precision of
DON measurements is of great concern. This
document covers the testing procedures used
by the Grain Inspection, Packers and Stockyards
Administration (GIPSA) of the U.S. Department
Fungi are probably one of the most numerous
plant families on earth. By definition they are
plants that contain no chlorophyll (can grow
in conditions of little or no natural light) and
range from single cells to a body of branched
hyphae (tubular filaments) that often produce
fruiting bodies that form molds, mushrooms,
smuts and yeasts. Instead of producing their
own food, fungi absorb nutrients from either
a living or dead host material. Mycotoxins
are metabolites (by-products) of the growth
of the molds.
is the parent fungi of
DON. Wheat and barley are the most commonly
effected grain crops but the same fungus
does infect corn. In the field, it shows up
as a brown discoloration at the base of barley
glumes, a pink to reddish mold on the glumes
and kernels of the wheat heads and the tips of
the ears of corn. Spores of the fungi can stay
dormant on infected residues left on or in the
The optimal temperature range for the DON
mold is 70 to 85 F with moisture levels preferred
to be greater than 20 percent. This particular
fungi has two distinct growth cycles,
with the mold growing during the warm temperatures
of daytime, while the toxins are produced
during the cooler temperatures of the
The first step in DON analysis is obtaining a
representative portion. Great care should be
taken when sampling, since sampling error can
be a significant source of variation.
Testing Trucklots of Barley and Wheat
for Deoxynivalenol (DON)
Obtaining a representative sample from a lot
of grain is an important and essential part of
mycotoxin analysis. If the sample is not representative,
the analysis result will not represent
the true quality of the lot. In order for a
sample to be considered representative, it must
1. Obtained with equipment/procedures
designed to obtain sample from all areas
of the lot;
2. Of appropriate size;
3. Adequately identified;
4. Handled in such a way as to maintain
A 1998 GIPSA study
of ten trucklots of DON contaminated barley
found evidence that some stratification of
DON may occur. The lots in the study were
from the 1997 harvest that were stored in
farmer’s bins over the winter. Some blending
of the barley is likely to occur as the barley is
moved from the field to farm storage and then
to commercial elevator. Stratification may be
more pronounced in lots coming directly from
the field at harvest. Truck lots may be stratified
in varying degrees depending on the nonuniformity
in the field and harvesting practices.
To ensure that the sample that is tested is accurate,
proper sampling techniques must be
used to obtain a representative sample. A “coffee
can” sample from the exposed layer of
grain in a hopper car or truck, or a “bucket”
sample as a truck or railcar is unloaded does
not give a representative sample of the lot as a
A large percentage of grain, as it travels from
the farm to the final consumer, is sampled with
a probe sometimes referred to as a trier. The
probe is the only sampling method approved
by GIPSA for stationary lots. If probe sampling
is performed correctly, the samples
drawn are considered representative.
Hand probes are constructed of brass or aluminum
and come in various sizes with standard
lengths of 5, 6, 8, 10, and 12 feet. The
type of carrier and depth of grain dictates
which probe length is used. For flatbed trucks
or trailers use a 5 or 6 foot probe. For hopper
bottom trailers 6, 8, or 10 foot probes are recommended.
Probe-type mechanical sampling systems
have replaced hand probes at many facilities.
GIPSA has approved two designs (gravity-fill
and core) for probe-type mechanical samplers.
In-load suction probes (air probe) are not approved
because they draw air through the
grain and vacuum excessive amounts of fine
material into the sample.
Figure 1. Grain probe or trier
Probe Core Probe
Figure 2. Mechanical Probe Designs
How the sample is obtained
with the probe will also affect the accuracy
of results. A study conducted by
Michigan State University found that the variability
of DON measurements in trucks of
newly harvested soft red winter wheat was significantly
higher if less than four probes were
taken from the lot.
GIPSA has established a sampling pattern for
each type of carrier. The sampling patterns
are designed to obtain a representative sample
of approximately 2000-2500 grams which is
more than adequate for DON analyisis.
The following diagrams indicate the standard
sampling patterns. Insert the probe at the
points marked, with the tip of the probe angled
ten degrees in the direction of the arrow.
When two arrows are shown, the tip of the
probe may be pointed in either of the indicated
directions at the sampler’s discretion.
The ground portion size required by GIPSA
for barley and wheat is approximately 100
grams. A 1998 GIPSA study of DON contaminated
barley has shown that increasing the
size of the portion ground does not appear to
significantly decrease the variability of DON
results in barley. This does not mean that
sample size is unimportant for DON analysis.
As sample size is decreased below 100 grams,
at some undetermined point, size would become
a significant factor.
Clean the sample to remove dockage using a
Carter dockage tester. The Carter dockage
tester uses aspiration (air) and a combination
of riddles and sieves to remove readily separable
Figure 3. Seven probe pattern for flat-bottom
trucks or trailers containing grain more than four
Figure 4. Nine probe pattern for flat-bottom trucks
or trailers containing grain less than four feet deep.
Figure 5. Four probe pattern for hopper bottom
trailers or containers.
Photo 1. Carter dockage tester