The effect of tapping for frankincense on sexual reproduction in Boswellia papyrifera

Summary

Study 1

In the Horn of Africa, frankincense (an aromatic wood resin) is obtained by tapping Boswellia papyrifera trees. However, production is declining as a result of poor Boswellia regeneration, possibly due to low production of viable seeds as a result of intensive tapping which reduces resources for flowering and seed production. This study investigated the problem of low seed production in wild-growing B.papyrifera. Two experiments were carried out in 1998-99 at sites with annually tapped trees and with trees not tapped for several years. In the first experiment (described here) the quality and germination of seeds from five sites were compared.

Study sites: The research was carried out in five Boswellia woodland areas (Adi-Ketina, Atawen, Ferhen, Molki and Shilalo) in southwest Eritrea. The sites were selected based on the occurrence of relatively intact B.papyrifera stands, the type of agro-ecological zone and accessibility. Sites differed in the number of years that trees had been tapped or tapping had been abandoned.

Resin tapping: Tapping starts at the beginning of the dry season (September) and coincides with the Boswellia sexual reproductive cycle. Tappers make incisions in the trunk bark with an axe. In the selected study sites, six cuts are made at points along the tree stem, the 'normal' intensity. Each tree is revisited at least seven times per harvest season. Tapping is discontinued during the rains.

Seed quality and germination: The effect of tapping on B.papyrifera seed quality and seed germination was determined at each site. Fresh seeds were collected from 10 to 15 adult trees. In the laboratory, a sample of 200 seeds was used to assess seed quality  (infected, empty and filled seeds). The number of infected seeds was determined visually. The remaining healthy-looking seeds were immersed in water to separate filled, sunken and empty, floating seeds. The proportions of insect-damaged seeds was assessed.

Germination success was determined in a subsample of 200 non-infected and filled seeds. Seeds were soaked in water for 12 h (considered to promote germination) then placed on Petri dishes (50 per dish) on moist filter paper, and maintained in an incubator. Each day, germinated seeds were counted and removed.

Seed viability and germination: The percentage of insect-damaged seeds was similar at all sites and ranged from 19% in non-tapped trees at Atawen to 25% in annually tapped trees at Shilalo. The number of filled seeds was up to three times higher in non-tapped trees (66 and 54%) than in annually tapped trees (19-22%). Seeds from all five sites started to germinate after about 5 days. Seeds with the highest germination success were from the non-tapped trees at Atawen (94%) and Adi-Ketina (80%), whereas those from annually tapped trees had lowest germination (Molki 16%, Ferhen 14.5% and Shilalo 13.5%).

Conclusions and recommendations: The frankincense tapping regime of B.papyrifera i.e. six tapping points per tree through the dry season, causes tree exhaustion and a decline in vitality. The results of this study indicate low production of seeds, and mainly non-viable ones, in annually tapped trees. This may be contributing to lack of natural regeneration of this treein Eri trea. To redress the decline in seed production, the authors suggest that less intensive tapping regimes should be introduced, with fewer tapping points. There should be years in which there is no resin harvesting so that trees can recover and replenish carbohydate reserves. In order to allow successful recruitment of seedlings, it is also considered that there should be period of no livestock grazing.

Study 2

In the Horn of Africa, frankincense (an aromatic wood resin) is obtained by tapping Boswellia papyrifera trees. In parts of the world, frankincense is of great economic and social importance being an important element of incense and perfumes. However, production is declining as a result of poor Boswellia regeneration, possibly due to low production of viable seeds as a result of intensive tapping which favours allocation of resources for synthesis of resin at the expense of flowering and seed production. This study is the first to look into the problem of low seed production in wild-growing B.papyrifera.

Two experiments were carried out in 1998–99 at sites with annually tapped trees and with trees not tapped for several years. In the experiment described here, the effects of three tapping regimes on flowering, seed production and seed germination in differentsized trees were investigated in three woodlands. In the other experiment (see Case 593) the overall quality and germination of seeds from five woodland sites were compared.

Study sites: The effect of tapping treatment on Boswellia papyrifera flowering and seed production (in different sized trees) was determined at three sites with distinctly different tapping histories: Atawen (14 years without tapping); Adi-Ketina (4 years without tapping); and Ferhen (tapped every year), in south-western Eritrea. The study sites were selected based on the occurrence of relatively intact B.papyrifera stands, the type of agro-ecological zone and accessibility. The region experiences a long dry season from mid-September to mid-June with a little rain commencing at the end of March.

Study species: Adult Boswellia papyrifera tree height ranges from about 6 to 12 m. After leaf fall in September, floral buds develop and flowering takes place in November and December. Fruits mature in January and February, and new leaves develop at the beginning of the small rains in late March. Tapping starts at the beginning of the dry season, before the emergence of floral buds, and coincides with the whole Boswellia sexual reproductive cycle.

Frankincense is obtained from the living bark (i.e. phloem). Local tappers make incisions in the trunk bark with a traditional 'mngaf' axe. In the selected study sites, six cuts are made at points along the tree stem, the 'normal' intensity. The resin that exudes is left to solidify into small lumps and collected every 3 weeks, when cuts are re-opened to promote resin flow. Each tree is revisited at least seven times per harvest season. Tapping is discontinued during the rains as yields are low and resin is washed away.

The effect of tapping treatment on flowering and seed production: At each site, 15 trees were selected in diameter at breast height (d.b.h.) classes of 10 (small), 20 (intermediate) and 30 cm (large trees). Five trees in each class were subjected to one of the following treatments: no tapping, tapping at six points (the tapping system normally used) and tapping at 12 points (a tapping system locally considered as 'heavy'). Tapping incisions were made with a mngaf on the east- and west-facing sides of the trunk. Before wounds were re-opened, solidified resin was removed. The experiment started in October 1998 and lasted for about 7 months, until the onset of the first rains in April 1999.

During the experiment, the formation of sexual reproductive structures was monitored. The number of inflorescences was counted per tree. The total number of fruits per tree was estimated by taking five randomly selected inflorescences, of which four were positioned at the periphery and one in the centre of the tree crown. In the laboratory, the number of fruits per inflorescence was counted. The total number of seeds per tree was estimated assuming an average of three seeds per fruit. For each tree, the length and width of individual fruits (n = 30) was measured and the average seed mass was determined for a bulked sample of 100 seeds.

Germination success: Germination success of seeds was determined only for Atawen, as only at this site did trees provided a sufficient amount of healthy looking and filled seeds (n = 200). In the laboratory, seeds were placed on Petri dishes, 50 per dish, with moist filter paper, and stored in an incubator at 20/30 °C (12:12 h) and 80% relative humidity. Moisture levels were maintained by watering. Every day, germinated seeds (radical emerged) were counted and removed.

Effect of tapping and  tree size on flowering and seed production: Analysis revealed a significant d.b.h., tapping and interaction effect (d.b.h. × tapping treatment) on the number of inflorescences, the number of fruits per inflorescence and the number of seeds. The effects of d.b.h. and tapping on seed mass were also significant. Reproductive variables of different sized trees responded differently to the tapping treatments.

Overall, the number of inflorescences, the number of fruits per inflorescence, the number of seeds and the seed mass decreased with increasing tapping irrespective of tree size at all sites, except at Ferhen where heavily and normally tapped trees had the same number of inflorescences. Sexual reproduction variables were highest in large trees and decreased with decreasing tree size except at Adi-Ketina. Here however, although the number of inflorescences of large trees was about twice that of intermediate trees, the number of fruits per inflorescence in large trees was about half that of intermediate ones. Consequently, at Adi-Ketina seed production was similar between large and intermediate trees. At all sites, large trees tended to produce slightly heavier seeds than small trees, with a trend towards heavier seeds at Atawen and Adi-Ketina than at Ferhen, particularly for non-tapped trees.

At Atawen and Ferhen, and to a lesser extent at Adi-Ketina, the effects of tapping regime were stronger in large trees than in intermediate and small ones e.g. at Atawen, transition from normal to heavy tapping resulted in a proportionally greater reduction of the mean number of inflorescences in large trees (34%) than intermediate and small trees (about 15%). Likewise, transition from no tapping to normal tapping resulted in a stronger decrease in the number of fruits per inflorescence in large trees (47%) than in small trees (19%), and seed production was more than halved (53%) in large trees whereas small trees had a reduction of 37%. Overall, transition from no tapping to heavy tapping resulted in around a 75% decline seed numbers at all sites (range: Atawen 69–76%, Adi-Ketina 75–76% and Ferhen 77–82%).

The number of inflorescences, the number of fruits per inflorescence and seed mass were consistently higher at the sites with a break from tapping (Atawen and Adi-Ketina) than at Ferhen, where trees are tapped every year. Particularly in large trees, a 14-year non-tapping period proved more beneficial than a 4-year period. Also, in general, tree size and non-tapping periods had a positive effect on fruit dimensions (i.e. they were bigger) and tapping a negative effect.

The effects of tree size and tapping on seed germination were only investigated for trees in Atawen, as this was the only site where individual trees yielded sufficient seeds (200/tree) for a germination test. For all tree sizes, germination success of seeds from non-tapped trees (90%) was significantly higher than those from normally (60%) or heavily tapped (40%) trees.

Conclusions and recommendations: The current frankincense tapping regime i.e. six tapping points per tree, causes tree exhaustion and a decline in vitality. This results in reduced flower and fruit production, and low production of viable seeds. To redress, it is suggested that less intensive tapping regimes should be introduced: the number of tapping points per tree should be reduced and there should be years in which resin harvesting is not undertaken so that trees can replenish their carbohydrate reserves. In this study, at a site with 4 years of no tapping, germination levels greatly increased. In order to allow successful recruitment of seedlings, it is considered that seed germination must be followed by a period of no livestock grazing.


Note: If using or referring to this published study, please read and quote the original paper. Please do not quote as a www.conservationevidence.com case as this is for previously unpublished work only. The original paper can be viewed at: http://blackwellpublishing.com/submit.asp?ref=0021-8901