Sampling Guidelines for Hemp
U.S. Domestic Hemp Production Program
Issued January 15, 2021
Purpose:
1. Standard and Performance-based sampling guidelines are specified for field and indoor
sampling of hemp. States and Tribes shall develop their own sampling protocols in accordance
with §990.3.
2. Samples are taken to obtain specimens for the measurement of total tetrahydrocannabinol
(THC) content, which determine whether the specimens are hemp or marijuana. The
measurements are intended to be representative of the total THC content in a “lot” of hemp crop
acreage as identified by the producer. Hemp producers may not harvest hemp prior to the hemp
being sampled for THC concentration. Testing procedures are provided in a separate guidance
document.
Scope:
1. Samples collected under this procedure are acceptable for submission to a qualified testing
laboratory for determination of total THC concentration in hemp. After December 31, 202
4, all
laboratories testing hemp under the U.S. Domestic Hemp Production Program must be registered
with the DEA in accordance with §990.3(a)(3)(iii)(H) and §990.25(g)(iii).
2. Since the THC content of hemp generally peaks as the plant ripens, the timing of when
sampling occurs is important to accurately measure total THC concentration and monitor
co
mpliance with the USDA hemp production program. Harvest shall be completed within 30
days from sample collection.
3. Samples shall be co
llected only by a trained sampling agent. Sampling agents must be trained
under applicable USDA, State, or Tribal training procedures. States and Tribes must maintain
information, available to producers, about trained sampling agen
ts. Hemp producers may not act
as sampling agents.
4. It is
the responsibility of the licensed producer to pay any fees associated with sampling.
5. I
t is the responsibility of the sampling agent to pay any fees associated with sampling agent
training or testing.
Summary of Practice:
1. This practice provides procedures for entering a growing area and collecting the minimum
number of plant specimens necessary to represent a homogeneous composition of the “lot” that is
to be sampled. A trained sampling agent enters a growing area, strategically examines the
growing area, establishes an approach for navigating the growing area, and collects individual
specimens of plants in order to obtain a representative sample of hemp in the designated lot.
2. Cuttings from each “lot” of hemp crop acreage, as identified by the producer, and submitted to
and uniquely identified by the Farm Service Agency (FSA) per the requirements of the USDA
hemp production program, shall be organized as composite samples. The terminology used by
FSA to denote land areas include terms like “farm,” “tract,” “field,” and “subfield,” which are
equivalent to AMS’s term “lot.” For the purposes of these procedures, a “lot” is a contiguous
area in a field, greenhouse, or indoor growing structure containing the same variety or strain of
cannabis throughout. In addition, “lot” refers to the batch of contiguous, homogeneous whole of
a product being sold to a single buyer at a single time. The size of the “Lot” is determined by the
producer in terms of farm location and field acreage and is to be reported as such to the FSA.
Performance-Based Sampling Protocols:
1. States and Tribes may develop performance-based sampling protocols.
2. Performance-based sampling protocols may consider seed certification processes, other
process that identify varieties that have consistently resulted in compliant hemp plants, whether
the producer is conducting research on hemp at an institution of higher learning or that is funded
by a Federal, State, or Tribal government, whether a producer has consistently produced
compliant hemp plants over an extended period of time, and other similar factors.
3. Performance-based sampling protocols may consider alternative requirements for operations
that grow “immature” cannabis that does not reach the flowering stage. These facilities may
grow seedlings, clones, microgreens, or other non-flowering cannabis, as determined by the State
or Tribe.
4. A performance-based sampling protocol must have the potential to ensure, at a confidence
level of 95 percent, that the cannabis plants will not test above the acceptable hemp THC level of
0.3 percent on a dry weight basis.
5. Regardless of the specific performance-based sampling requirements developed under a State
or Tribal plan, all samples must be collected from the flowering tops of the plant by cutting the
top five to eight inches from the “main stem” (that includes the leaves and flowers), “terminal
bud” (that occurs at the end of a stem), ”or “central cola” (cut stem that could develop into a bud)
of the flowering top of the plant.
6. States and Tribes are required to include performance-based sampling protocols in the plan
submitted to USDA for approval if they decide to use this methodology.
Standard Sampling Protocols:
1. The standard sampling method must be used by all producers, except for producers operating
under a State or Tribal plan that includes a performance-based sampling requirement.
2. The standard sampling protocol ensures, at a confidence level of 95 percent, that no more than
one percent of the plants in each lot would exceed the acceptable hemp THC level and ensures
that a collected sample represents a homogeneous composition of the lot.
3. Every lot and every producer must be sampled and tested.
4. All samples must be collected from the flowering tops of the plant by cutting the top five to
eight inches from the “main stem” (that includes the leaves and flowers), “terminal bud” (that
occurs at the end of a stem), or “central cola” (cut stem that could develop into a bud) of the
flowering top of the plant.
5. All producers licensed directly by USDA are subject to these requirements.
Equipment and Supplies:
1. Garden pruners/shears (Cleaned prior to and following each composite sample. Some
examples of appropriate cleaning agents and supplies to use on garden pruners/shears are bleach,
rubbing alcohol, steel wool, and/or sandpaper.)
2. Sample bags, paper.
2.1. The size of the bags will depend upon the number of clippings collected per lot.
2.2 The bags should be made from material known to be free from THC.
3. Security tape
4. Permanent markers
5. Sample collection forms
6. GPS Unit of lot being sampled
7. Disposable gloves – Nitrile
8. Ladder
Sampling Guidelines:
1. The licensee or designated employee should be present throughout the sampling process, if
possible.
2. Surveillance of the growing area.
2.1. The sampling agent should estimate the average height, appearance, approximate density,
condition of the plants, and degree of maturity of the inflorescences (flowers/buds).
2.2. The sampling agent should visually establish the homogeneity of the stand to establish that
the growing area is of like variety.
3. Time of Sampling:
3.1. Within 30 days prior to the anticipated harvest of a designated hemp lot, an approved
sampling agent, State or Tribally designated person or Federal, State, local, or Tribal law
enforcement agency shall collect representative samples from such cannabis plants for THC
concentration level testing.
4. Field Sampling:
4.1. The licensee or designated employee should accompany the sampling agent throughout
the sampling process, if possible.
5. Surveillance of the growing area.
5.1 The sampling agent should verify the GPS coordinates of the growing area as compared
with the GPS coordinates submitted by the licensee to USDA.
5.2 The sampling agent should estimate the average height, appearance, approximate density,
condition of the plants, and degree of maturity of the inflorescences (flowers/buds).
5.3 The sampling agent should visually establish the homogeneity of the stand to establish
that the growing area is of like variety.
6. Time of Sampling:
6.1 Within 30 days prior to the anticipated harvest of a lot a sampling agent should collect
representative samples from such a lot for THC concentration level testing.
7. Field Sampling:
7.1 For purposes of determining the number of individual plants to select for sampling, the size
of the growing area should be considered. For sampling purposes, samples from separate lots
must be kept separate and not be comingled.
7.2 For lots of less than one acre, including greenhouses, select a minimum of 1 plant, then take
a cutting from the plant to form a sample. For lots of 1 to 10 acres, including greenhouses, follow
the chart in example 2 below, take cuttings of each plant, then combine to form a composite
sample.
7.3 For growing areas larger than ten (10) acres, including greenhouses, the number of plants
that should be selected to form a composite sample is based upon the Codex Alimentarius
Recommended Methods of Sampling for the Determination of Pesticide Residues for
Compliance with MRLS CAC/GL 33-1999.
7.4 The sample size is estimated in a two-step process. The first step is to estimate the
number of primary plants to be sampled. The second step is to adjust the estimate of primary
plants by the acreage under cultivation.
The initial number of primary plants is estimated using:
where p is the confidence level to detect hemp plants testing above the acceptable THC threshold
and i is the proportion of hemp plants having THC content above the acceptable threshold. The
values for i are based on past experience in the same or similar growing areas and should be
consistent with the requirements currently in the Final Rule.
The initial primary plants estimate is adjusted by the number of acres to calculate the minimum
number of primary plants as follows:
where n is the minimum number of primary plants to be selected for forming a composite
sample, n
0
is the initial number of primary plants estimated using the previous formula, and N is
the number of acres under cultivation.
Examples 1 and 2 below describe the minimum number of samples that must be collected in
order to meet the 95% confidence level requirements in the Final Rule. If a State or Tribal hemp
program does not have data from a prior growing season to determine the i value, the sampling
charts below may be utilized. State and Tribal hemp programs are free to include more rigorous
sampling requirements, or to develop performance-based requirements.
Example 1: The initial primary plant sample size is 299 with a confidence level of 95% to detect
hemp plants having an acceptable hemp THC level and a proportion of hemp plants having THC
content above the acceptable threshold equal to 0.01 is considered appropriate. The adjusted
primary plant sample sizes for fields from 11 to 173 acres in size are shown in the following
table:
:-
,-
:
:
:
:-
i-
:-
i-
:-
c
:
:-
:-
i-
:-
i-
-
Number
of acres
Sample
Size
11
11
12
12
13
13
14
14
15
15
16
16
17
17
18-
19
18
20
19
21
20
22
21
23
22
24
23
25-
26
24
27
25
28
26
29
27
30
28
31-
32
29
33
30
34
31
35
32
36
33
"'
37-38
34
39
35
Number
of acres
Sample
Size
40
36
41-
42
37
43
38
44
39
45-
46
40
47
41
48
42
49-
50
43
51
44
52
45
53-
54
46
55
47
56
48
57-
58
49
59
50
60-
61
51
62
52
63-
64
53
65
54
66-
67
55
68
56
69-
70
57
71
58
72-
73
59
74
60
Number
of acres
Sample
Size
75-
76
61
77
62
78-
79
63
80-
81
64
82
65
83-
84
66
85-
86
67
87
68
88-
89
69
90-
91
70
92
71
93-
94
72
95-
96
73
97-
98
74
99
75
100-
101
76
102-
103
77
104-
105
78
106-
107
79
108
80
109-
110
81
111-
112
82
113-
114
83
115-
116
84
117-
118
85
Number
of acres
Sample
Size
119-
120
86
121-
122
87
123-
124
88
125-
126
89
127-
128
90
129-
130
91
131-
132
92
133-
134
93
135-
136
94
137-
138
95
139-
140
96
141-
143
97
144-
145
98
146-
147
99
148-
149
100
150-
152
101
153-
154
102
155-
156
103
157-
157
104
159-
161
105
162-
163
106
164-
166
107
167-
168
108
169-
170
109
171-
173
110
Example 2: The adjusted primary plant sample sizes for fields from less than 1 to 10 acres in size
are shown in the following table:
7.5 Sampling agents should always walk at right angles to the rows of plants if possible,
beginning at one point of the lot and walking towards another point on the opposite side of the
lot. If the lot is too dense for this to be possible, the sampling agent should take all reasonable
steps to ensure that a sample is collected that represents a homogeneous composition of the lot
by avoiding edges and thoroughfares.
7.6 While walking through the growing area, the sampling agent should cut at least “n”
inflorescences (the flower or bud of a plant) based on the acreage of the growing area, at random
but convenient distances. Avoid collecting sample specimens from the borders of the
field/greenhouse.
7.7 The cut should be obtained from the flowering tops of plants when flowering tops are
present and shall be approximately five to eight inches in length from the “main stem” (that
includes the leaves and flowers), “terminal bud” (that occurs at the end of a stem), or “central
cola” (cut stem that develops into a bud) of the flowering top of the plant.
7.8. Utilize paper sample bag(s) for collecting sample cuttings. Ensure that each bag has the
minimum number of cuttings, n, as calculated by 7.4, or in the Example Tables 1 and 2. If one bag
cannot accommodate the minimum number of cuttings due to lot size, the sample may be divided
into multiple bags, but must be clearly labeled in such a way that each bag is appropriately matched
with the corresponding lot. (i.e. For lot 101 with three corresponding sample bags: 101 1 of 3, 101 2
of 3, 101 3 of 3.)
7.9. Seal each bag and record the sample number or other documentation as required by the State or
Tribe.
7.10
A sampling protocol must have the potential to ensure, at a confidence level of 95 percent,
that the cannabis plants will not test above the acceptable hemp THC level of 0.3 percent on a
dry weight basis.
8. Sample identification:
8.1 The sampling agent should seal each bag and record the sample identification number. The
sample should also be identified with the following information: Sampling agent contact information;
name and contact information of the producer; producer hemp license or authorization number; date
of sample; and lot, subfield, or other identifier as provided by the USDA Farm Service Agency; any
other information that may be required by States, Tribes, Law Enforcement Authorities, mail
delivery services, customers, or groups of customers.
Note: In accordance with 7 CFR 1.901(e), the contents of this document do not have the force and effect
of law and are not meant to bind the public in any way, and the document is intended only to provide
clarity to the public regarding existing requirements under the law or agency policies.
