vol 6 no 2 previous article vol 6 no 2 table of contents vol 6 no 2 next article

Industrial hemp growth trials - New Brunswick 1998

Charles Schom

Surge Inc., 157 Water St., Unit "C", Saint Andrews, New Brunswick E5B 1A7, Canada

Hemp is a very versatile plant. Its seed contains all of the essential human dietary amino acids as well as the two essential fatty acids, and it tastes good! The stalk produces three sizes of fiber, a long fiber ideal for textiles and twine, a medium fiber about the length of softwood fiber and a short fiber about the length of hardwood fiber. Further, the chemical components are ideal as raw materials for the manufacture of Rayon and cellophane (Bócsa and Karus 1997).

Hemp was grown in the province of New Brunswick (NB) in Canada in the first half of the century. Hemp production declined in Canada through the 1930's. Cotton and wool processing was mechanized while hemp processing remained dependent on manual labor. This was coupled with the fear that marijuana might corrupt Canadian youth, which led to laws prohibiting the growth of hemp in 1938 (Twomey 1995). The law was set aside in 1941 as fiber was needed for the war effort, but enforced once the war need ended. It was again grown in Canada by non-government agencies under research licenses in 1995, increasing from 10 licenses in 1995 to 31 in 1997 (Health Canada 1997). As of April 1, 1998 new legislation was enacted by the Canadian federal government, allowing the commercial cultivation of industrial hemp. This legislation sets a minimum size of 4 hectares for commercial production. Thus, any smaller area must be grown under a research license.

Industrial hemp missed the rigorous agronomic, genetic and biological assessment other crop species in North America were subjected to between 1943 and 1990, as little was grown legally in North America during that period. In trials at experimental stations across Canada, including Fredericton, NB, growth and fiber production were adequate, but seed set was poor at Fredericton (Gehl 1995). Unfortunately, the strains used in those trials no longer exist, so there was no way of knowing, a priori, exactly how the industrial hemp strains available today would perform in New Brunswick. Further, as it has not been farmed since the 1940's, few farmers in New Brunswick had experience working with it as a crop. Thus, the need for a modern re-assessment of hemp in New Brunswick was indicated.

There were a several objectives of this study. The first was to develop a pool of farmers with experience growing industrial hemp. The second was to assess the ability of available farm equipment to successfully cultivate and harvest industrial hemp. The third objective was to test a broad a range of sites and soil conditions to determine which would grow a good hemp crop. The fourth was to test cultivars against each other versus the New Brunswick environment, weather, disease, etc.. The fifth was to assess the impact of fertilizer on growth. Finally, the ability and conditions following harvest needed to be assessed to determine if field retting would work in New Brunswick.

Four cultivars were compared: two Hungarian fiber varieties ('Kompolti' and 'Uniko-B') and two Romanian seed varieties ('Secuieni-1' and 'Irene'). Each was grown on a quarter acre which, in turn, was divided into three fertilizer treatments. The design was repeated ten times, i.e., run at ten sites around New Brunswick. Height and dry weight yield correlated closely; therefore, height was used to describe growth. The experiment was set up to conform to a four way factorial analysis of variance. No statistically significant differences existed between cultivars or sites. Sowing date and fertilizer application did have statistically significant differences. The failure to identify cultivar differences overall is probably due to variability within each variety. However, when comparisons were made within fertilizer treatments, differences could be identified. A number of questions that were raised, but not answered in this study, center on the availability of nutrients and interactions with moisture. A drought-induced calcium deficiency seems to have killed the meristematic region of a majority all plants in all of the cultivars at one site. The yield (dry matter per acre) for the full fertilizer treatment did vary between sites, but was over 4,000 pounds per acre at one site and this in just a little over 70 days after planting. However, there was no significant seed crop. The maturing heads were attacked by a fungus, Botrytis cinerea which prevented normal maturation. It remains to be seen if this is the result of an unusual year. There were differences between cultivars as to the number of plants infected and the severity of the infection, implying that some genetic variation in resistance may exist.

A number of conclusions were drawn. Some of them are based on statistically significant differences, others are observational. All of them need to the tested to confirm or reject the contention raised. This data provides a bench mark to be both challenged and act as a starting point for the future. The conclusions fall into three overlapping categories: Agronomic, Biological and Mechanical.


1) Emergence rate is critical to optimizing return. There does seem to be a cultivar specific response to the amount of fertilizer applied. The full fertilizer treatment (120 pounds of nitrogen, 100 pounds of potassium and 150 pounds of phosphate) was too much for 'Kompolti' and 'Uniko-B', but not 'Secuieni-1'.

2) The 'Irene' grown in a trial which included a double dose of fertilizer produced more dry matter than those grown in the full fertilizer treatment. The 'Secuieni-1' was the same. The 'Kompolti' and 'Uniko-B' full fertilizer treatment had higher yields than the double fertilizer. This may have been in part because the emergence rate was down in the double fertilizer treatment.

3) Seeder type that rolls seed in gave lower emergence rates than seed drills, whether mechanical or air powered.

4) Any kind of soil compaction seemed to reduce emergence and early growth rate.

5) The pesticide MCPA drifted onto the hemp field and damaged plants. None of the chemicals used on potato fields showed any signs of damaging the hemp.

6) The relationship between soil analysis and availability of inorganic elements to industrial hemp is not clear. The addition of urea alone at a site with good soil analysis did not improve the yield as much as adding the full fertilizer did. In all cases the full fertilizer treatment out performed the half fertilizer treatment. The control plot always had the lowest yield.

7) The test plot that the province of New Brunswick planted out-performed our project at that same site. The province applied less fertilizer initially, then followed with a second application. Even though they planted nearly 2 weeks later, their final yield was higher.

8) Organic practices did not give yields comparable to the full fertilizer treatment. Organic approaches should be further investigated.

9) Yields on the order of 3-4 tons per acre should be regularly achieved given a 90 day growing period.

10) Retting by leaving the cut hemp in the field, did not seem to go to completion in the normally recommended 4 weeks. The harvested stock which seemed best retted had been in the field for two months.

11) Top-killing the plant, leaving them stand to ret then cut and separate, is an approach that should be explored.


1) As seed ages, the percent that germinate decline and fewer of those that do germinate, successfully emerge.

2) Percentage long fiber produced varies from cultivar to cultivar and from site to site.

3) Industrial hemp is capable of producing a large root mass. However, it does not normally do so because its roots are not very aggressive. If nutrients are in some way limited and there is an open space, the root will expand into it and the plants grow taller than their neighbors, so long as they are protected from the wind. If exposed to the wind or disturbed at a regular interval, the root growth appears to be reduced and the plants are smaller than their neighbors.

4) No site produced yields of grain at a level anywhere near what would be considered commercially viable. All sites suffered as a result of fungal infection. As there are cultivar differences in susceptibility to fungus, resistant strains should be developed.

5) There appears to be a relationship between increased incidence of fungal Botrytis infection, and increased percentage of THC, the psychoactive agent, in the industrial hemp plant.

6) Drought in combination with poor soil, will reduce growth, but not kill the plants.

7) A drought-induced calcium deficiency can kill the growing tip but not the plants themselves. Most of the affected plants produced new growth and seed.

8) Insects inhabit the crop, bees harvest the pollen, sucking insects feed off the plants. Lady bugs also reside on the hemp.

9) Animals seem to begin to avoid hemp fields not very long after planting.

10) Industrial hemp has a strong odor which changes at least twice, with the three different odors associated with a different stage of the plants life cycle.

11) Industrial hemp may produce chemicals with bacteriastatic action. as the author both cut himself and developed severe blisters while working in the hemp fields. At no time did he see any signs of infection, and experienced relatively little pain and far more rapid healing than expected.

12) While there was a great deal of difference in the times that the hemp was planted, all plants within a cultivar were very close to the same stage of sexual maturity at all sites. This includes the site planted July 20. The photoperiod response that triggers sexual maturation is very strong, leading to the diversion of energy away from growth and into reproduction.


1) Most planters, cutters and harvesters can be used with industrial hemp. Some work better than others.

2) Exposed shafts, gears or other devices that turn will become wrapped with the long fiber. It accumulates in corners and on chains.

3) Working small areas, such as this project dealt with, presents no real problem other than a longer time commitment to use the equipment than might be expected.

4) Equipment has to be disassembled and cleaned. Fiber has to be picked out of the chain drives and cut off the shafts.

5) The rule seems to be: "make modifications that reduce the chance of the fiber wrapping, stop often to check the equipment, clear it and go slowly."


This project would not have been possible without the support and/or hard work of a number of people. First and foremost are the 10 farmers who joined with me in growing the crop. (Even those that did not follow the protocol, contributed to our understanding of industrial hemp.) The farmers were Lee Ann and Larry McConnell, Sean McBrine, Barb and Phillip Roach, Ivan Nyborg, Harry Pond, Ben Baldwin, Dwayne Hicks, Mary Thompson, Phil Moore and Sandy Muir.

The other underpinning of the project was the Province of New Brunswick through both its Strategic Partners Grants Program and the staff. I cannot say enough about David Walker and the help he and his people gave. He was always ready to find some time in his busy schedule to discuss the results and help with the interpretation, advising on ways of dealing with problems.


Bócsa, I and M. Karus 1997. The Cultivation of Hemp. Hemptech, Sebastopol, California.

Gehl, D. 1995. Summary of Hemp Research in Canada. Agriculture and Agri-Foods Canada, Indian Head, Saskatchewan.

Health Canada 1997. Summary Report on Hemp Research Licenses (1995-1997). www.hc-sc.gc.ca

Twomey, S. 1995. The re-emergence of fiber hemp in Canada. Global Biodiversity 5(4): 9-11.

vol 6 no 2 previous article vol 6 no 2 table of contents vol 6 no 2 next article