32


Genetic future of hemp

David P. Watson and Robert C. Clarke

International Hemp Association
Postbus 75007, 1070 AA Amsterdam, The Netherlands


        What will be the traits required for future industrial hemp varieties?  Many industrial hemp cultivators will desire increased cellulose content for biomass fuel production, higher primary fiber yields for pulping, or high yields of extra fine fibers for textiles.  Seed growers and processors will require high-yielding varieties with increased levels of fatty acids or other target compounds.  Included in this list would also be common agronomic traits such as tolerance to drought, wet heavy soils, salt, cold, heat or humidity and resistance to pests and diseases.  The general agronomic suitability of hemp varieties for specialized altitudes or latitudes beyond the limited present-day European range of industrial hemp growing is also a particularly important future consideration.

Table 1. Registered industrial hemp varieties (47 total) and their commercial availability (21 total).

  France - 10
Available:
     'Fedora 19'
     'Felina 34'
     'Fedrina 74'
     'Futura 77'
Unavailable:
     'Fibrimon 21'
     'Fibrimon 24'
     'Fibrimon 56'
     'Férimon 12'
     'Epsilon 68'
     'Santhica 23'
Hungary - 5
Available:
     'Kompolti'
     'Kompolti Sárgasázrú'
     'Uniko-B'
     'Kompolti Hibrid TC'
     'Fibriko'
Poland - 3
Available:
     'Bialobrzeskie'
     'Beniko'
Unavailable:
     'Dolnoslaskie'
Romania - 4
Available:
     'Fibramulta 151'
     'Lovrin 110'
     'Sequeni'
     'Irene'
Bulgaria - 2
Unavailable:
     'Mecnaja copt.'
     'Silistrensi'
Ukraine - 11
Available:
     'YUSO 13'
     'YUSO 31'
     'Glukhov 33'
     'Zenica'
Unavaillable:
     'Dneprovskaya
        Odnodomnaya 6'
     'Kuban'
     'YUSO 11'
     'YUSO 14'
     'YUSO 15'
     'YUSO 16'
Italy - 5
Unavailable:
     'Carmagnola'
     'Carmagnola
        Selezionata' or 'CS'
     'Fibranova'
     'Eletta Campana'
     'Suprafibra'
Slovenia - 2
Unavailable:
     'Rudnik'
     'Pesnica'
Spain - 2
Unavailable:
     'Delta 405'
     'Delta-Llosa'
Czechoslovakia - 1
Unavailable:
     'Rastslaviska'
Yugoslavia - 1
Available:
     'Novosadska Konoplja'
Germany - 1
Available:
     'Fasamo'
 

        No developed hemp varieties exist that are suitable for equatorial or even subtropical latitudes.  Of the 47 hemp varieties registered or in commercial trade (Table 1), European varieties account for all of them.   All were developed in (and for) regions north of the 45th parallel and in general will not perform well if moved closer to the equator by even as little as 10-15 degrees, as demonstrated by recent field trials in South Africa (Dippenaar et al. 1996).  The limited production of hemp crops in subtropical Asia relies solely on unimproved landraces of variable cannabinoid content.  The International Hemp Association (IHA) often receives requests for low-THC, EU-registered industrial hemp varieties that would be suitable for Brazil, Cambodia, Jamaica, Hawaii, Nigeria, Viet Nam, The Congo (formerly Zaire) and other tropical and sub-tropical regions.  We are forced to respond that none of the available registered varieties will perform well in those regions.   Northern temperate varieties are adapted to long summer daylength and flower in Europe in the Autumn, when the days become shorter.  In equatorial regions, the daylength is never long enough to prevent flowering and these northern varieties only reach a meter or two in height before they flower.  European varieties growing in foreign environments are also attacked by a wide range of pests and diseases for which they have no resistance.  We suggest that there is an immediate need for a tropical hemp collection and equatorial facilities to be established for tropical industrial hemp variety development.  Currently, only unimproved landraces (with fiber content below 20% and THC levels that may reach 3%) satisfy the daylength restrictions of tropical areas.
        Canadian and northern European farmers have the opposite problem with the available selection of European cultivars.  Few European varieties will mature early enough to allow seed production and variety multiplication in the climatic and daylength conditions of Canada and northern Europe.  In 1925, R. J. Hutchinson, the Canadian Chief Officer of the Division of Economic Fiber Production observed:

“A considerable handicap to the growing of hemp in Canada is the difficulty of obtaining good seed at a reasonable price.  While our climate is favorable for producing hemp for fiber, the growing season, in most districts, is not long enough to fully mature the seed.  Subsequently, hemp seed production trials were undertaken near Ottawa and other stations in areas with longer growing seasons: Summerland, B.C.; Sidney, B.C.; Sanichton, B.C.; Harrow, Ontario; and Kentville, Nova Scotia.  These experiments demonstrated that the only areas in Canada with a growing season suitable to reliably mature seed of such a high fiber yielding variety as Minnesota No. 8 [extinct American cultivar] was the southern tip of the Niagara Peninsula near Harrow.”

This situation has not changed for Canada in the last 70 years.
        Confusion in the public’s mind between “rope and dope” continues to plague the industrial hemp industry in many parts of the world.  Producers of recently registered monoecious varieties with trace or claimed null levels of THC, such as the newest Ukrainian ‘YUSO-15’ or the French ‘Epsilon 68’ and ‘Santhica 23’ can only hope to solve the “THC problem” for a limited portion of hemp’s potential range.  While it is easier to reduce the THC content of monoecious varieties through in-breeding, growing these types may suffer serious drawbacks, e.g., decreased vigor due to inbreeding depression resulting in lowered fiber and seed yield.  The present European Union (EU) THC limit of 0.3% is certainly low enough to prevent any diversion of industrial hemp to drug purposes and no further reductions are necessary.  However, as a strategic economic and political maneuver, the French may use their newest varieties to push for unneeded EU legislation that would forbid the cultivation of any industrial hemp variety containing more than 0.1% THC, mimicking regulations in the former Soviet Union.   This legislation would, in effect, outlaw all registered varieties except those of the French and a few Ukrainian varieties.
        The EU recently adopted a resolution that subsidies will be paid only to hemp growers who do not harvest their hemp until its seeds have ripened (see pg. 38).  If the stand is harvested before the seeds are ripe, the farmer will not only be refused the subsidy, but may also be liable for a 50,000 DM fine (Bócsa 1996)!  This resolution is especially damaging to the southern Italian varieties, and may prevent the registration of high yielding dioecious varieties such as ‘Kompolti Hibrid TC’ and ‘Novosadska Konoplja’ because they take longer to mature than the French varieties.   However, any farmer north of France trying to allow the crop to mature its seed in the often cold, wet and foggy Autumn harvest weather will encounter difficulties in most years.  The fact that harvesting hemp crops to seed maturity compromises fiber quality also seems to have escaped the EU regulators.
        To date, there is only one hemp variety (Uniko-B) specifically designed for high-yield seed production.  One to 1.5 t/ha is presently the maximum possible seed yield using this high yielding Hungarian unisexual variety.  Seed yields from the dioecious variety ‘Novosadska Konoplja’ in Yugoslavia also approach these levels.  Additional high-yielding seed varieties must be developed if hemp seed is to compete in price with other grains.  With a seed variety developed for high yield and mechanized harvesting, the yields might be greatly increased.  Drug varieties developed for high flower yield have produced more than a kilogram of seed per plant.  Cultivars also will be developed for the food industry with specific desirable protein and oil profiles.  The nutritional value and implications of hemp seed have recently been reviewed by Deferne and Pate (1996).
        Interest in hemp food varieties is increasing, as shown by Theimer and Mölleken (1997) in their world survey of seed fatty acid and protein profiles (see pg. 13).  They are examining the entire Vavilov Research Institute (VIR) collection of more than 400 Cannabis accessions, the 45 currently registered hemp varieties and about 100 landraces from around the world.   Canada’s $60,000 research grant to Ruth Shamai, President of The Natural Order, will kick-start a North American research into hemp seed and oil production.   The pioneering work of Callaway and Laakkonen (1996) in Finland may accelerate this process.  Callaway et al. (1996) announced a new fatty acid discovery that reveals the nutritional superiority of, and possible “nutriceutical” applications for, high-latitude hemp seed suitable for cultivation across Canada and the Nordic countries.   This break-through would not have been possible without VIR 313 and VIR 315, the initial breeding materials that originated from the VIR Cannabis germplasm collection.   Other genetically determined seed traits to be explored may include special amino acid profiles, better balanced essential fatty acid (EFA) contents for food use, or unusual oil components for industrial uses.
        Hemp varieties with extremely small seeds for baking, very large seeds for easier hulling and thin and soft-shelled variants for snack foods, are possible to develop for whole seed use.  The variability of hemp seed size is striking.  Industrial hemp varieties normally produce seeds of 40-60 seeds per gram, but wild Cannabis growing at 3000 meters in Kashmir produces tiny seeds of more than 1000 to the gram.  Extremely large seeds of 15-25 to the gram from Afghanistan, Chile, China, Japan, Korea and Pakistan have been reported.  One of the largest expenses in hemp fiber growing is seed cost.  If hemp seeds were half their current weight, then farmers could buy and sow half as many kilograms of seed to achieve the same crop density.
        Varieties could be developed with gross morphological markers to help agricultural inspectors identify approved industrial varieties.  Suggestions have included red, yellow or purple plants, those having leaves with different shapes or with even numbers of leaflets, and varieties with THC genes that could be turned off or disabled permanently.  Hemp with yellow stems has been developed for paper (e.g., ‘Kompolti Sárgaszárú’) thereby producing a better color paper using less bleach during pulping.  It has been suggested that the use of ‘Kompolti Sárgaszárú’ as a five percent admixture to approved commercial hemp seed would facilitate field recognition of industrial hemp crops by simply observing the color of the stalks (Bócsa 1994).
        Late-maturing bulky hemp varieties with little fiber, but high total cellulose yield might better suit biomass projects.  Dr. James Burke has grown hemp in Carlow, Ireland for four years to determine the effectiveness of burning hemp for electrical energy production, rather than to continue to rely solely on rapidly dwindling peat reserves.  In addition, he has announced a contest for the best power plant design to use this new fuel.  The power plant will run on hemp, paper wastes, and chicken manure and may be operational as soon as 1999, supplying power to 30,000 homes, a significant percentage of Irish electrical needs.  A cultivar should be developed to suit the needs of such biomass energy producers.
        Pharmaceutical varieties that are rich in cannabinoids are currently being developed that can potentially relieve suffering for millions worldwide.  Natural THC can be grown and extracted at a fraction of the cost of the synthetic equivalent and will successfully compete in the legitimate cannabinoid pharmaceutical market, currently valued at over US$ 20 million.  A host of other naturally occurring cannabinoids (e.g. CBD) could also be approved for a wide range of indications and varieties would be developed to supply any of these cannabinoids in high concentrations.
        Modifying or increasing the secondary metabolites produced by Cannabis may help promote its resistance against attack by pests.   Unfortunately, we have even heard of considerations for incorporating herbicide or pesticide resistance into industrial hemp varieties, similar to that found in soybean varieties genetically engineered to be resistant to the broadleaf herbicide, Roundup®.   We believe that this would be tragic.  We must be careful not to create the next agro-nightmare, as is cotton, the largest user of farm chemicals of any crop.   Hemp already offers us an environmentally compatible fiber, seed and medicinal crop that is free of a dependence upon agricultural chemicals.  There is nothing “wrong” with hemp and there is no need to “fix” it.  Breeders need only develop cultivars for specific uses and regions of cultivation.
        Questions arise as to who will develop these desired new varieties and where will they obtain the initial genetic raw materials required to develop them?  The 46 currently registered varieties were developed primarily by starting with a few promising local landraces selected over hundreds of years by peasant farmers in areas of traditional hemp cultivation.  The “modern” industrial hemp varieties are descendant from very few ancestors.  The European varieties are based on only three gene pools: Northern and Central European ecotypes, Southern European ecotypes and East Asian ecotypes (de Meijer 1995).  The 33 registered hemp varieties for which pedigree information is available consist of 23 monoecious, 14 dioecious and one unisex types.  All originated entirely, or in part, from landraces of the Central and Southern European ecotypes and only two of the Hungarian varieties incorporate ancestors from the Far Eastern ecotype.  All of the monoecious and unisex varieties derived their monoecious trait from ‘Fibrimon’.  Seven of the nine dioecious varieties include genes from Hungarian landraces and varieties that were all originally derived from Italian landraces.  A single accession from China was used to establish the hybrid “triple cross” varieties from Hungary.   Italian landraces, combined with the single monoecious line ‘Fibrimon’ and a single Chinese landrace (along with a few other Central and Southern European accessions) were used to breed all current industrial hemp varieties (Table 2).  This is an extremely narrow gene pool for an entire crop plant and explains why industrial hemp varieties are so poorly suited to growing in regions outside Europe.

Table 2. Country of origin, ancestors, geographical origins and ecotype/gene pool classification of the currently available registered hemp varieties (de Maijer 1995).

  Country/
Sexual type
Ancestors Geographical
Origins
Ecotype/
Gene pool
 
  France
     10 monoecious
'Fibrimon'

with exotic
combinations from

Russia
Italy/Turkey
Russia,
Hungary and Italy
Central European
Southern European
Central European
Southern European
 
  Hungary
     4 dioecious
     1 unisex
Fleischmann
'Kompolti'
'Fibrimon'

Kinai

Hungary/Italy
Hungary/Italy
Russia
Italy/Turkey
China
Southern European
Southern European
Central European
Southern European
Far Eastern
 
  Poland
     3 monoecious
'Kompolti'
'Fibrimon'

'LKCSD'
'Bredemann 18'

Italy
Russia
Italy/Turkey
Russia
Russia
Southern European
Central European
Southern European
Central European
Central European
 
  Romania
     2 dioecious
     2 monoecious
'ICAR 42-118'
Silistra landrace
'Fibrimon'
Italy/Turkey
Bulgaria
Russia
Italy/Turkey
Southern European
Southern European
Central European
Southern European
 
  Ukraine
     2 dioecious
     7 monoecious
'Szegedi'
'Fibrimon'
Hungary
Russia
Italy/Turkey
Southern European
Central European
Southern European
 
  Yugoslavia
     1 dioecious
Fleischmann Hungary/Italy Southern European  
  Germany
     1 monoecious
'Bernburg' and
'Schurigs'
Russia Central European  
  Italy 'Carmagnola'
'Bredemann Elite'
Italy
Russia
Southern European
Central European
 

        Cannabis most likely originated in Central Asia and soon spread to Europe, India and China, following ancient trade routes, and eventually reached Africa and the New World.  Local landraces evolved in all of these introduced climates and new cultural settings.  Until the 1970s, many indigenous cultures still used hemp and continued to grow their own local landraces each year.  Today, few traditional cultures still grow and process hemp.  Almost all have stopped using hemp and now prefer to buy market cloth or spin and weave flax, cotton, wool or synthetics.  Increased availability, lower cost, ease of use, and better performance for some applications like wool (for warmth) and synthetics (for maritime uses) have pushed hemp into relative obscurity in a short time.  When a local landrace is not reproduced every five to ten years, the seeds will most likely die and that germplasm may be gone forever.  Seeds must be properly stored in a gene bank and reproduced periodically under ideal conditions.  The vast majority of landraces may already be extinct simply because of neglect resulting from modern market pressures.
        Another obstacle for breeders intending to work with industrial hemp is the confused legal status of the genus Cannabis.  In many countries, even in traditional hemp cultivation areas, all Cannabis is illegal to grow, regardless of its intended end use.  In some regions, even hemp seed is illegal.  Plants with little THC content are often destroyed even when found growing wild, because they look like marijuana.  The last 60-70 years have been disastrous for the Cannabis gene pool, and many local landraces, the result of hundreds of years of selection for local use, have been lost because of Cannabis eradication, neglect on the part of agricultural officials and industry, anti-Cannabis propaganda and the general trend (until recently) to reduce industrial hemp breeding and research.  Genetic materials are a living heri-tage and we are their custodians.   We must concentrate our efforts to collect, preserve, characterize and utilize the remaining Cannabis genetic resources before it is too late.
        As the worldwide reduction in diversity of local hemp landraces used by indigenous farmers continues, the importance of gene banks becomes obvious.  Unfortunately, few comprehensive Cannabis germplasm collections exist.  Most of them are held by national gene banks that may or may not share their valuable inventories with other breeders.  The largest collection of hemp germplasm is maintained by the Vavilov Institute of Plant Research (VIR) in St. Petersburg, Russia, and, with a cooperative development agreement, they will share small amounts of varieties for which they have sufficient seed.
        In 1992, in order to preserve the VIR’s collection for future breeders, the International Hemp Association (IHA) negotiated a humanitarian aid project, the VIR/IHA Cannabis Germplasm Preservation Project, to ensure the timely reproduction and survival of the entire VIR Cannabis seed collection.  The project work began in 1993.  From 1993 to 1996, the IHA has donated US$ 50,000 for project funding.  Unfortunately, because of the disjunctions associated with the changeover from a centralized command economy to a capitalist market system, science in general and the VIR in particular, have suffered tremendously.   Three quarters of the VIR staff of 2000 have been laid off, salaries are up to 6 months late, and there is no funding to reproduce the 400,000 seed accessions of other economic plants and their wild relatives held by the largest and oldest gene bank in the world!
        In 1993, the IHA/VIR project developed two key criteria be fulfilled before a Cannabis accession was considered adequately reproduced.  The primary goal of the germplasm preservation project is the conservation of the entire gene pool of each accession.  It is very important that the population size is large enough to ensure that as many of the genes as possible within the accession’s gene pool are reproduced in the resultant seed.  A minimum limit of 1,000 plants was set for monoecious varieties, and 2,000 plants for dioecious ones.   This assures that 99% of the gene pool will be reproduced with each reproduction (Crossa et al. 1993).  Unfortunately the seed reserves of many of the accessions consisted of less than 1,000 viable seeds, making our goal impossible to achieve.  The secondary goal is to reproduce the accessions in sufficient quantities to maintain a reserve for future reproductions and to distribute seed to researchers worldwide.  A minimum limit of 300 grams of seed was set for long term storage and later reproductions.  Seeds can be released to the research community only if more than this amount is held by the VIR.  Three hundred grams is approximately 15,000 seeds and will allow the following three storage regimes:

1.)  5,000 seeds stored at an ambient temperature of 15 degrees Celsius and moisture content of about 10%, in an active collection for reproduction,
2.)  5,000 seeds kept at 4-6 degrees Celsius and 7% moisture content in refrigerated medium-term storage, and
3.)  5,000 seeds archived in a reserve collection at -20 degrees Celsius and 6% moisture in frozen long-term storage.

About 40 percent of the seed accessions are now stored in refrigerated or frozen storage.
        The IHA, with the generous support of our members, has nearly saved the VIR’s Cannabis germplasm collection from extinction, but to use this resource to its full potential, the entire collection must be grown out in common gardens at appropriate latitudes and evaluated for uniqueness, variety purity and the relative value of its agronomic traits.  The VIR made a limited evaluation of the collection in 1975 and the results were published by the VIR in Russian, but this effort should be repeated and expanded from a modern breeder’s perspective.   Breeders worldwide could then more easily understand and utilize this valuable resource.
        Because of the technical and financial difficulties in reproducing the VIR Cannabis collection, there has likely been considerable loss of genetic diversity and purity through low population sizes and incomplete isolation.  In addition, many of their accessions may be so similar to each other that they need no longer be represented and reproduced as separate accessions, which only results in pointless extra maintenance, storage and reproduction costs.   VIR materials are also present in limited amounts in other gene banks and research institutes, where they are adequately maintained and stored.
        A common goal of hemp breeders should be to establish a smaller and broader core collection of Cannabis germplasm that has been exhaustively characterized agronomically, in the field and at the molecular and chemical levels in the laboratory.  Only then can we see what diversity really is available for future research.  This core collection could then be maintained with optimal reproduction and storage methodology.  The individual accession evaluations should also be presented in a concise form making it accessible to a wide range of plant breeders.
        Unfortunately, the 1997 VIR/IHA Cannabis Germplasm Preservation Project is currently slowed due to insufficient funding.   While most of the collection is not presently at risk, the planned grow-out and evaluation of the entire collection will not happen this year.  Anyone who can help locate sources of funding for next year’s VIR project is invited to contact the IHA.

References


BACK HOME NEXT