Hemp (Cannabis sativa L.) is a day-length sensitive, short-day plant, meaning it begins to flower once daylight hours drop below a certain threshold, known as the critical photoperiod. The latitude where a hemp variety was developed strongly influences this threshold. Varieties bred in northern regions typically have longer critical photoperiods—an adaptation to long summer days—than those developed farther south (Hall et al. 2012). When grown for fiber, the most desirable varieties produce tall, consistent stems (at least 6 feet) before the onset of flowering. Once the crop flowers, stem elongation slows and lignin deposition and secondary fiber production increases, resulting in a lower-quality fiber, specifically for textiles (Liu et al. 2015).
At the start of the hemp pilot program in 2014, fiber variety options were limited. Most commercially available varieties were imported from European countries such as France, Italy, and Poland. For instance, the variety ‘Bialobrzeskie’ was developed in Poznań, Poland (52.40° N)—a latitude similar to southern Canada (Figure 1). Summer days in Poland are much longer than those in North Carolina; the longest day of the year in North Carolina is about 14 hours on June 23, which corresponds to a daylength typical of late August in Poland (Figure 1).
As a result of relatively shorter summer days, ‘Bialobrzeskie’ flowered much earlier under North Carolina conditions and produced short stems. These early-flowering European varieties quickly proved unsuitable for fiber production in the region because they did not remain vegetative long enough to generate profitable stem biomass.
To identify varieties better adapted to local conditions, we evaluated fiber hemp from Kunming, China, in 2021 and 2022 (Halker et al. 2024). These lines, originating from a more southerly latitude (Figure 1), flowered in early September and produced substantial biomass, with stems reaching 8 to 10 feet in height.
Since conducting those trials in 2021 and 2022, the number of commercially available fiber hemp varieties has increased considerably. The objective of this study was to establish a fiber hemp variety trial to evaluate these newer lines and determine their suitability for textile fiber production systems in North Carolina.
Figure 1. Comparison of latitudes for the US, Europe, China, and their day lengths.
Materials and Methods
Small plot field trials were conducted in 2023 and 2024. Trials were planted at the Caswell Research Station in Kinston, North Carolina (2023 and 2024), and the Cherry Research Station in Goldsboro, North Carolina (2024). Ten varieties were evaluated in 2023 and nine in 2024. All varieties were treated as fiber only (harvested at male flower maturity) except for Ursa ‘Grande’, which was bred for and treated as dual purpose (producing fiber and seed). Trials were arranged in a randomized complete block design with three blocks per location. One week prior to planting, we applied a pre-emergent herbicide (2.5 pints per acre of Ethalfluralin [Sonalan® HFP]) and a granular fertilizer (75 pounds of nitrogen per acre [lb N/ac] of 46-0-0 SuperU™). Planting occurred during the first week of June with a small-plot drill at a rate of 1.5 million pure live seed per acre (PLS/ac), except for Ursa ‘Grande’, which was planted at 600,000 PLS/ac. An additional 75 lb N/ac of granular fertilizer was applied at canopy closure.
Once male flowers began shedding pollen, we noted the date, collected a flower sample for THC compliance testing, and harvested two 1 square meter (m2) subsamples of stem material per plot. Ursa ‘Grande’ subsamples were not collected until about 75% of mature seed had formed. We measured the length and diameter of ten representative stems per subsample and left the material to ret in the field for approximately four weeks, or until the bast easily separated from the hurd. We then dried the material to measure straw yield and ran it through a 16-roller decorticator to generate bast yields. Bast fiber samples were further processed and opened using hand carders. A 10 gram (g) subsample of the opened fiber was chemically degummed and analyzed for physical quality (Figure 2).
Figure 2. Pictorial diagram of the different stages of fiber hemp stem processing.
Results
2023
Harvest Time
Harvest time for fiber-only varieties ranged from late July (AV1 and Jurassic) to the end of September (GVA-H-23-1142). Ursa ‘Grande’ was not harvested until October 9 to allow for seed maturity (Table 1).
Stem Length and Width
Ursa ‘Grande’ produced significantly longer and thicker stems than all other varieties. No differences were observed among the remaining varieties, though there was a numerical trend for increased length as harvest date was delayed. Stem width did not show this same trend (Table 1).
| Variety | Company | Purpose | Harvest Date | Length (cm) |
Width (mm) |
Total THC (%) |
Straw Yield (lb/ac) |
Bast Yield (lb/ac) |
Percent Bast (%) |
|---|---|---|---|---|---|---|---|---|---|
|
AV1 |
Acquiflow |
Fiber |
7/28 |
168 b* |
5.56 b |
0.07 b |
4,495 d |
1,124 b |
25.22 a |
|
Ursa ‘Grande’ |
Cornell |
Dual |
10/9 |
264 a |
11.18 a |
7,914 abcd |
1,362 ab |
17.20 c |
|
|
GVA-H-23-1142 |
Cornell |
Fiber |
9/28 |
202 b |
6.96 b |
0.14 ab |
9,195 ab |
2,177 a |
23.83 ab |
|
GVA-H-23-1143 |
Cornell |
Fiber |
9/1 |
211 b |
6.89 b |
0.03 b |
8,880 abc |
2,094 ab |
23.33 abc |
|
Ursa ‘Alta’ |
Cornell |
Fiber |
9/21 |
194 b |
6.35 b |
ND |
9,820 a |
1,853 ab |
18.75 bc |
|
IH Williams |
IND HEMP |
Fiber |
8/14 |
185 b |
6.90 b |
0.22 ab |
6,733 abcd |
1,538 ab |
22.97 abc |
|
Yuma S2 |
Kanda Hemp |
Fiber |
8/3 |
194 b |
6.62 b |
0.26 ab |
5,482 bcd |
1,214 ab |
23.68 abc |
|
Yuma TS2 |
Kanda Hemp |
Fiber |
8/3 |
198 b |
7.08 b |
0.34 a |
5,149 cd |
1,181 ab |
23.02 abc |
|
FIBER |
Kanda Hemp |
Fiber |
8/3 |
204 b |
5.90 b |
0.05 b |
5,008 d |
1,329 ab |
26.46 a |
|
Jurassic |
Verve Seeds |
Fiber |
7/28 |
188 b |
7.34 b |
0.06 b |
4,440 d |
1,141 ab |
25.99 a |
*Sharing the same letter within a column indicates that the values are not significantly different (p > 0.05). ↲
ND = Nondetectable ↲
Total THC
Yuma TS2 tested above the United State Department of Agriculture (USDA) total THC compliance threshold of 0.30%. GBA-H-23-1142, IH Williams, and Yuma S2 were below the 0.30% total THC threshold but not significantly different from Yuma TS2. The remaining varieties produced either a nondetectable amount of THC or were close to 0.00%.
Straw Yield, Bast Yield, and Percent Bast
Straw yields varied dramatically among varieties. Ursa ‘Alta’ produced the most straw (9,820 pounds per acres [lb/ac]) while Jurassic (4,440 lb/ac) and AV1 (4,495 lb/ac) produced the least. Straw yield generally followed the same trend as stem height: Later harvest date resulted in increased straw yield.
Differences in bast yield were not as dramatic. GVA-H-23-1142 produced the most bast (2,177 lb/ac) but was only significantly more than one variety, AV1 (1,124 lb/ac), which produced the least. All remaining varieties had intermediate bast yields.
Percent bast (calculated as bast yield ÷ straw yield) also differed among varieties, and earlier maturing varieties (Jurassic, AV1, FIBER) tended to have higher percent bast content.
Degummed Fiber Quality
No differences in the amount of weight lost after degumming were observed among varieties (caustic loss; Table 2). The amount of fiber remaining after carding (carding yield) was lowest in Ursa ‘Grande’ (74.50%) and highest in IH Williams, Yuma S2, Yuma TS2, and FIBER (ranging from 93.70% to 94.20%). The diameter of these carded fibers (micron) again depended on variety, with IH Williams and FIBER producing the finest fiber (19.0 micrometer [μm]) and Jurassic producing the coarsest (22.4 μm). The variability in carded fiber diameter (micron standard deviation) differed among varieties; however, the differences were marginal: Jurassic had the highest micron standard deviation and IH Williams the lowest.
| Variety | Company | Purpose | Caustic Loss (%)† |
Carding Yield (%) |
Micron (μm) |
Micron Standard Deviation |
|---|---|---|---|---|---|---|
|
AV1 |
Acquiflow |
Fiber |
36.63 |
93.13 ab* |
20.9 abcd |
17.4 ab |
|
Ursa ‘Grande’ |
Cornell |
Dual |
36.03 |
74.50 e |
21.4 abc |
17.0 ab |
|
GVA-H-23-1142 |
Cornell |
Fiber |
37.57 |
87.17 abcd |
20.2 bcd |
16.8 ab |
|
GVA-H-23-1143 |
Cornell |
Fiber |
38.13 |
85.87 bcd |
22.2 ab |
17.7 ab |
|
Ursa ‘Alta’ |
Cornell |
Fiber |
37.67 |
85.57 cd |
21.6 abc |
17.8 ab |
|
IH Williams |
IND HEMP |
Fiber |
35.73 |
94.20 a |
19.0 cd |
16.5 b |
|
Yuma S2 |
Kanda Hemp |
Fiber |
36.90 |
93.87 a |
19.8 bcd |
16.9 ab |
|
Yuma TS2 |
Kanda Hemp |
Fiber |
34.50 |
93.60 a |
20.8 abcd |
17.4 ab |
|
FIBER |
Kanda Hemp |
Fiber |
36.53 |
93.70 a |
19.0 cd |
17.1 ab |
|
Jurassic |
Verve Seeds |
Fiber |
35.67 |
89.83 abc |
22.4 a |
18.3 a |
*Sharing the same letter within a column indicates that the values are not significantly different (p > 0.05). ↲
†No significant differences among varieties for caustic loss. ↲
2024
The 2024 season was challenging. Significant drought affected most of June and July, which was then followed by heavy rainfall from Tropical Storm Debby and Hurricane Helene. Though the crop survived, growth was negatively impacted. Furthermore, plots were lost in Goldsboro due to standing water. These lost plots ultimately reduced statistical power, and thus minimal differences were observed among varieties.
Harvest Time
Harvest timing in 2024 (Table 3) was comparable to what we observed in 2023 (Table 1). The earliest harvest was for AV1 and FIBER 1 at the end of July and the latest harvest was at the beginning of October for Ursa ‘Grande’ and GVA-H-23-1142.
Stem Length and Width
Ursa ‘Grande’ produced the longest stems, which were significantly longer than AV1 in Kinston, and all other varieties were intermediate. No significant differences in length were observed among varieties in Goldsboro, though the numerical trends follow those seen in Kinston. Ursa ‘Alta’ and AV1 produced the thinnest stems in Kinston and Ursa ‘Grande’ produced the thickest. Again, no significant differences in width were observed among varieties in Goldsboro.
Total THC
None of the varieties produced total THC above the 0.30% threshold. The highest total THC was produced by ZL1 in Goldsboro (0.20%).
| Variety | Company | Purpose | Harvest Date* | Length (cm) |
Width (mm) |
Total THC (%) |
|||
|---|---|---|---|---|---|---|---|---|---|
| Kinston | Goldsboro | Kinston | Goldsboro | Kinston | Goldsboro | ||||
|
AV1 |
Acqui Flow |
Fiber |
7/31 |
138.22 b |
103.71 |
4.87 b‡ |
4.63 |
0.12 |
0.05 |
|
Ursa ‘Grande’ |
Cornell |
Dual |
10/2 |
196.32 a |
186.72 |
8.45 a |
7.49 |
0.15 |
|
|
GVA-H-23-1142 |
Cornell |
Fiber |
10/2 |
172.33 ab |
142.74 |
5.54 ab |
7.11 |
0.01 |
0.04 |
|
GVA-H-24-1380 |
Cornell |
Fiber |
8/26 |
174.78 ab |
161.99 |
5.40 ab |
6.41 |
0.07 |
ND |
|
Ursa ‘Alta’ |
Cornell |
Fiber |
9/25 |
156.43 ab |
183.56 |
4.12 b |
6.52 |
0.04 |
ND |
|
IH Williams |
IND HEMP |
Fiber |
8/13 |
172.52 ab |
154.27 |
5.46 ab |
7.27 |
0.14 |
0.17 |
|
Yuma |
Kanda Hemp |
Fiber |
9/10 |
180.07 ab |
157.94 |
5.91 ab |
7.74 |
0.04 |
ND |
|
ZL 1 |
Kanda Hemp |
Fiber |
9/10 |
167.77 ab |
5.41 ab |
† |
0.09 |
0.20 |
|
|
FIBER 1 |
Kanda Hemp |
Fiber |
7/31 |
177.97 ab |
168.56 |
6.05 ab |
5.92 |
0.12 |
0.15 |
|
P-value |
|
0.0406 |
0.3892 |
0.0265 |
0.5859 |
0.2584 |
0.4714 |
||
*Harvest date for Kinston. Goldsboro was harvested +/- 3 days from Kinston. ↲
†ZL 1 plots were destroyed by standing water in Goldsboro. ↲
‡Sharing the same letter within a column indicates that the values are not significantly different (p > 0.05). ↲
ND = Nondetectable. ↲
Straw and Bast Yields
Straw yields in Kinston ranged from 2,840 lb/ac (AV1) to 8,379 lb/ac (GVA-H-23-1142; Table 4). Yields were generally lower in Goldsboro but followed a similar pattern of differences among varieties seen in Kinston. However, Ursa ‘Alta’ produced 8,643 lb/ac of straw in Goldsboro, which was the highest yield achieved for the season.
Bast yields in Kinston differed among varieties. Variety GVA-H-23-1142 produced the most bast in Kinston (2,647 lb/ac), which was significantly higher than all other varieties in that location. Minimal differences in bast yield were observed among varieties in Goldsboro; however, Ursa ‘Alta’ and GVA-H-23-1142 produced significantly more bast than Yuma, IH Williams, GVA-H-24-1380, and AV1.
Percent bast ranged dramatically in Kinston from 4.77% (GVA-H-24-1380) to 31.70% (GVA-H-23-1142). GVA-H-23-1142 produced nearly identical percent bast (31.56%) in Goldsboro as it did in Kinston. More broadly, percent bast results tended to be similar for varieties across locations. The one exception is FIBER 1, which produced higher percent bast in Goldsboro (25.41%) than Kinston (17.74%).
| Variety | Company | Purpose | Straw Yield (lb/ac) | Bast Yield (lb/ac) | Percent Bast (%) |
|||
|---|---|---|---|---|---|---|---|---|
| Kinston | Goldsboro | Kinston | Goldsboro | Kinston | Goldsboro | |||
|
AV1 |
Acqui Flow |
Fiber |
2,840 e* |
2,164 b |
409 d |
446 b |
14.24 cd |
18.61 bc |
|
Ursa ‘Grande’ |
Cornell |
Dual |
5,688 bcd |
4,885 ab |
1,442 bc |
1,249 ab |
25.31 ab |
25.63 ab |
|
GVA-H-23-1142 |
Cornell |
Fiber |
8,379 a |
6,357 ab |
2,647 a |
1,974 a |
31.70 a |
31.56 a |
|
GVA-H-24-1380 |
Cornell |
Fiber |
6,379 abcd |
4,940 ab |
301 d |
602 b |
4.77 d |
11.94 c |
|
Ursa ‘Alta’ |
Cornell |
Fiber |
7,346 abc |
8,643 a |
1,576 b |
2,275 a |
21.24 bc |
26.31 ab |
|
IH Williams |
IND HEMP |
Fiber |
4,952 cd |
3,323 b |
1,294 bc |
758 b |
26.63 ab |
22.34 b |
|
Yuma |
Kanda Hemp |
Fiber |
7,346 abc |
4,650 ab |
1,059 bcd |
569 b |
14.37 cd |
11.76 c |
|
ZL 1 |
Kanda Hemp |
Fiber |
8,015 ab |
1,420 bc |
† |
17.74 bc |
† |
|
|
FIBER 1 |
Kanda Hemp |
Fiber |
4,186 de |
3,703 ab |
687 cd |
937 b |
15.24 c |
25.41 ab |
|
P-value |
|
<0.0001 |
0.0340 |
<0.0001 |
0.0008 |
<0.0001 |
0.0007 |
|
†ZL 1 plots were destroyed by standing water in Goldsboro. ↲
*Sharing the same letter within a column indicates that the values are not significantly different (p > 0.05). ↲
Degummed Fiber Quality
Similar to 2023 (Table 2), no differences were observed among varieties for caustic loss in either 2024 location (Table 5).
No significant differences were observed among varieties in both locations for carding yield. Similar to 2023, in 2024 at Kinston, IH Williams produced the thinnest fiber, though differences among varieties in that location were minimal. No significant differences were observed for fiber micron or micron standard deviation in Goldsboro. Finally, IH Williams and Ursa ‘Grande’ produced fiber with the lowest micron standard deviation in Kinston. Ursa ‘Alta’ had the highest micron standard deviation in this location.
| Variety | Company
|
Purpose | Caustic loss (%)* | Carding yield (%) | Micron (μm) | Micron standard deviation | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Kinston | Goldsboro | Kinston | Goldsboro | Kinston | Goldsboro | Kinston | Goldsboro | |||
|
AV1 |
Acqui Flow |
Fiber |
36.4 |
36.5 |
89.8 |
91.9 |
20.0 ab* |
20.0 |
17.7 ab |
17.6 |
|
Ursa 'Grande' |
Cornell |
Dual |
34.2 |
36.0 |
89.7 |
80.7 |
19.8 ab |
19.3 |
16.9 b |
16.8 |
|
GVA-H-23-1142 |
Cornell |
Fiber |
37.6 |
33.0 |
89.8 |
90.1 |
19.4 ab |
18.0 |
17.1 ab |
16.4 |
|
GVA-H-24-1380 |
Cornell |
Fiber |
36.1 |
35.6 |
85.2 |
87.3 |
21.8 ab |
19.8 |
17.5 ab |
16.2 |
|
Ursa 'Alta' |
Cornell |
Fiber |
38.2 |
34.4 |
88.0 |
89.6 |
21.2 ab |
19.0 |
18.4 a |
16.2 |
|
IH Williams |
IND HEMP |
Fiber |
34.6 |
35.1 |
93.2 |
91.5 |
19.0 b |
18.9 |
16.7 b |
16.6 |
|
Yuma |
Kanda Hemp |
Fiber |
35.2 |
37.7 |
89.7 |
88.4 |
22.7 a |
19.3 |
17.6 ab |
16.6 |
|
ZL 1 |
Kanda Hemp |
Fiber |
34.2 |
88.9 |
† |
22.6 a |
† |
17.9 ab |
† |
|
|
FIBER 1 |
Kanda Hemp |
Fiber |
36.4 |
36.0 |
89.4 |
92.3 |
19.7 ab |
18.4 |
17.3 ab |
16.5 |
|
P-value |
|
|
0.2068 |
0.1368 |
0.1212 |
0.2171 |
0.0079 |
0.3136 |
0.0110 |
0.8481 |
*Sharing the same letter within a column indicates that the values are not significantly different (p > 0.05) ↲
†ZL 1 plots were destroyed by standing water in Goldsboro. ↲
Key Takeaways
Results from these trials demonstrate that several commercially available and precommercial fiber hemp lines exhibit late flowering and produce tall, slender stems with abundant biomass. In general, a longer vegetative growth period was associated with greater total biomass production. However, an extended growing season may also increase hurd content, thereby reducing the percentage of bast fiber.
Differences in the physical properties of degummed fibers were observed among varieties, though these differences were not always consistent. The variety IH Williams tended to produce finer fibers with a lower micron standard deviation, indicating more uniform fiber fineness.
Currently, no universal standards or grading system exist for fiber hemp, making it difficult to draw definitive conclusions about the relationship between variety and fiber quality. When contracting to grow fiber hemp, growers and processors should discuss the intended end-use of the crop, as this will largely determine which varieties are most suitable.
References
Hall, Jack, Surya P. Bhattarai, and David J. Midmore. 2012. “Review of Flowering Control in Industrial Hemp.” Journal of Natural Fibers 9 (1): 23-36. ↲
Halker, China Allissa P., Alex L. Woodley, S. Chris Reberg-Horton, Shannon Henriquez Inoa, and David H. Suchoff. 2024. “Evaluating Chinese Fiber Hemp (Cannabis sativa L.) Varieties and Planting Dates in North Carolina.” Agrosystems, Geosciences & Environment 7 (3): e20569. ↲
Liu, Ming, Dinesh Fernando, Geoffrey Daniel, Bo Madsen, Anne S. Meyer, Marcel Tutor Ale, and Anders Thygesen. 2015. “Effect of Harvest Time and Field Retting Duration on the Chemical Composition, Morphology and Mechanical Properties of Hemp Fibers.” Industrial Crops and Products 69: 29-39. ↲
Publication date: Dec. 15, 2025
AG-1000
N.C. Cooperative Extension prohibits discrimination and harassment regardless of age, color, disability, family and marital status, gender identity, national origin, political beliefs, race, religion, sex (including pregnancy), sexual orientation and veteran status.