Slide 1
Slide 2

Hemp variations as pulp source researched in the Netherlands

Posted on July 1, 1993

The feasibility of nonwood pulp production by means of hemp (Cannabis sativa L.) is currently under investigation in the Netherlands. Research ranging from breeding to pulp technology and market survey is carried out at several institutes of the Agricultural Research Dept. (DLO). This effort is part of a comprehensive search for profitable new nonfood crops for Dutch agriculture. This article reports on some results of the screening of a germ plasm collection for variation in characters relevant for the introduction of Cannabis as an arable pulp crop. Selected genotypes are presently used for breeding experiments (Figure 1). (Figure 1 omitted)

Establishment & Maintenance of the Collection

A collection of approximately 200 populations was established at Centre for Plant Breeding & Reproduction Research — Agricultural Research Dept. (CPRO-DLO) over the past five years.1 It comprises fiber and drug strains and wild or naturalized populations from a worldwide geographic origin ranging between 28° and 58° latitude.

Apart from cultivars of which seeds were commercially available, each population has been regenerated to obtain sufficient seeds for evaluation trials. Some research is being carried out to identify optimal conditions for long-term storage of the collection.

Evaluation Topics

The main objectives of the breeding research are an adaptation of stem quality to the requirements for pulp production and an increase of stem yield to make hemp competitive with other arable crops and traditional sources of pulp. Because of legal implications, contents of narcotic compounds in Cannabis need to be low. Furthermore, the crop should be a poor host for soil pathogenes. These topics are important in the germ plasm screening, besides various morphological, physiological, and biochemical traits for a general characterization of populations.

Stem Quality

The bark and woody core of dicotyledoneae possess distinct properties. For a detailed characterization of stem quality, these two fractions and their constituent fibers need to be discriminated. The bark fibers of Cannabis are the traditional fibers of commerce used for cordage and textiles. The woody core has usually been considered waste. The present research focuses on the utilization of entire stems. Therefore, the germ plasm evaluation covers bark as well as woody core properties.

Within the bark, a distinction is made between primary and secondary fibers. Properties of the distinguished stem fractions are summarized in Table 1. (Table 1 omitted) For bark, the chemical properties concern the i extracted fibers, whereas for woody core, the contents of the entire tissue are given.

When compared with conifer fibers, the woody core fibers of the hemp stem are much shorter, which has a negative effect on paper strength.2 The length of secondary bark fibers is comparable to that of conifer fibers. Primary bark fibers are much longer, which makes them suitable raw material for a range of high-quality paper grades. Because of its low cu-cellulose content, the potential utilization of the woody core seems restricted to mechanical pulps,3,4 whereas the bark of hemp is already used in chemical pulps for specialty papers.

The collection screening comprised the estimation of the mass fractions of primary and secondary bark fibers and woody core in the stem dry matter. For the woody core, dimensions of the fibers were also assessed. The laborious determination of chemical characters was omitted, as there seemed little prospect for finding variation among populations. Only small differences were previously found within a set of French and Italian hemp cultivars in contents of (alpha-cellulose, hemicellulose, and lignin measured in bark and wood separately.4

Figure 3 shows the variation in stem composition observed in field-grown stems of 92 populations of very distinct origin and domestication. (Figure 3 omitted) The scheme illustrates what hemp fiber breeding in this century has achieved. Total bark fiber content ranges from 12% in wild populations and drug strains to 281 in modern cultivars (x-axis). With increasing total bark fiber content, populations show a gradual decrease of the woody core fraction from 74% to 521, and an increase of the secondary and primary bark fiber fraction from 1% to 10l and from 10l to 23%, respectively. Selection for increased bark fiber content was discontinued beyond a level of about 30%. This was due to the inevitable increase of the relatively short and coarse secondary bark fibers that negatively affected rope and textile quality.

The average length of woody core fibers of 98 evaluated populations ranged from 484 mum to 607 mum. Resulting differences between populations were, however, not statistically significant. The average diameter of these fibers did differ significantly; they ranged from 25 mum to 38 mum. The lack of variation in woody core fiber length indicates that for paper pulp production, breeding should give priority to a continued replacement of woody core by bark fiber, rather than to the improvement of woody core properties.

Stem Production

In field experiments in the Netherlands, maximum yields were obtained from 14 to 16 metric tons of stem dry matter per hectare (one hectare = 2.47 acres). In Cannabis, there is large variation in dates of anthesis (full blooming) and seed maturity that depends primarily on origin latitude. Field-grown populations that are acclimatized to and grown in the Netherlands (52° latitude) flower at the end of June. Populations that are imported from higher latitudes flower earlier, and for those from lower latitudes, anthesis is delayed up until mid-September. In general, late-flowering populations have larger stem yields Figure 4. (Figure 4 omitted) Therefore, beside selection for vigorous growth, this variation in life-cycle duration enables an increase of stem production by simply organizing the production of seed for sowing at a much lower latitude than the latitude of cultivation.

Narcotic Compounds

The presence of narcotic compounds is generally considered an important reason for the decline of hemp cultivation in the course of this century. Outdoor screening of 97 populations showed significant variation in the average content of the cannabinoid THC, which ranged from 0.06 to 1.77 in the female inflorescence leaf dry matter.5 For comparison, THC contents exceeding 10 are not uncommon in marijuana produced by seedless clones of superior genotypes in greenhouses and growth chambers. Outdoors, in densely spaced crops, such contents will not occur, even in drug strains. Although high bark fiber content does not necessarily exclude high THC content, most fiber cultivars have very low THC content and thus possess no psychoactive potency.

Soil Pathogens

Although hemp itself is generally considered to tolerate continuous cultivation on the same field for periods of five to 10 years, host-characteristics for soil pathogenes are important with regard to other susceptible crops in the narrow rotations of the Netherlands. Reactions of Cannabis to the root-knot nematodes Meloidogyne hapla and M. chitwoodii are currently being studied and results indicate that Cannabis is a moderate host for M. hapla and a poor host for M. chitwoodii.6

One hundred and forty-eight populations of the Cannabis collection were screened for resistance to M. hapla. significant variation in host characteristics was found. The results of the test agreed sufficiently with nematode infection and multiplication on a naturally infested arable field, which indicates that there seem to be good possibilities for breeding highly resistant cultivars.


There is sufficient variation within the genus Cannabis for further genetic improvement of quality and yield of hemp fiber as a nonwood source of pulp. Fiber cultivars with negligible low contents of the narcotic compound THC are already commercially available. As a relatively low-input crop that can be grown at a wide range of latitudes, hemp seems very suitable for mass production of nonwood cellulose. The poor host characteristics to plant-parasitic nematodes, which can even be improved by breeding, make hemp also suitable for more intensive agricultural systems with narrow rotations of susceptible dicotyledonous crops. However, the main factors for a successful introduction of hemp as a pulp source are not botanical or agricultural, but industrial and political considerations.


  1. E.P.M. de Meijer and L.J.M. van Soest, 1992, The CPRO Cannabis germ plasm collection, Euphytica, 62: 201-211.
  2. I.M. Wood, 1982, The utilization of field crops and crop residues for paper pulp production, Field Crop Abstracts, 34: 557-568.
  3. A. Bosia, 1975, Hemp for refiner pulp, Poper, World Research, and Development Number 1975: 37-41.
  4. L. Triolo, 1980, Materie prime non legnose per l’industria cartaria, Italia Agricola, 1: 33-61.
  5. E.P.M. de Meijer, H.. van der Kamp and F.A. van Eeuwijk, 1992, Characterisation of Cannabis accessions (populations) with regard to cannabinoid content in relation to other plant characters, Euphytica, 62: 187-200.
  6. G.M.C. Coenen, personal communication, 2993.
  7. L.C. Anderson, 1974, A study of systematic wood anatomy in Cannabis, Botanical Museum Leaflets, Harvard Univ. 24: 29-36.
  8. D. Catling and . Grayson, 1982, Identification of vegetable fibres, Chapman & Hall, London — New York.
  9. O. Heuser, 1927, Hanf and Hartfaser, Julius Springer Verlag, Berlin. p. 156.
  10. W. Hoffmann, 1961, Hanf, Cannabis sativa. In: H. Kappert and W. Rudorf Eds. Handbuch der Pflanzenzuchtung. Band V. Paul Parey, Berlin-Hamburg. pp. 204-261.
  11. B.C. Kundu, 1942, The anatomy of two Indian fibre plants, Cannabis and Corchorus with special reference to fibre distribution and development, I. Indian Bot. Soc., 21: 93129.
  12. H.J. Nieschlag, G.H. Nelson, LA. Wolff, and R.E. Perdue, 1960, A Search for New Fiber Crops, TAPPI, 43: 193-201.

Dr. de Meijer is a research worker in the potential crops section of the Centre for PIant Breeding & Reproduction Research-Agricultural Research Dept. (CPRO-DLO), P.O. Box 16, NL 6700 AA Wageningen, the Netherlands.

Copyright © 1993, Pulp & Paper. All rights reserved.