Distribution and Abundance of the Edible Orchids of the Southern Highlands of Tanzania

All orchids have showy flowers and many are therefore popular as exotic houseplants as well as contributing significantly to eco-tourism. A special attribute of some orchid genera is the production of edible tubers, an attribute that can contradict eco-tourism if harvesting is not done sustainably. Such is the case in the Southern Highlands where for example, Kitulo Plateau has been gazetted to be a National Park but where edible orchid harvesting is a serious conservation problem. Botanically, unsustainable harvesting can mean the loss of important germplasm even before some species are known to exist. A field survey was undertaken for three weeks in March 2002 in the Southern Regions of Tanzania (Iringa, Mbeya, Rukwa and Ruvuma) to study aspects of the extent of the distribution, diversity and density of edible orchids. Tools for identification included structured questionnaire, on-the-spot identification as well as using herbarium voucher samples and keys. Analytical tools included Shannons’ diversity index and Canonical Community Ordination (CANOCO) and ANOVA in Microsoft Excel. A total of 13 edible species of orchids were documented during the survey all of them from the genera Habenaria, Disa and Satyrium. While Tandala and Molo hills had the highest species Diversity Index (over 1.2), Kitulo Plateau showed the greatest species density of over 200/30m quadrants. Local communities in Madaba in Ruvuma, Molo, and Mbisi in Rukwa regions used the orchids as vegetables and snacks in lieu of animal meat. The rest of the studied areas used orchids as a source of cash by selling to Zambian traders and or to their middlemen. More research is needed to monitor growth phases, reproductive/propagation studies as well as taxonomic and systematic studies preferably using molecular techniques to determine relationships amongst edible orchids and their non edible relatives. INTRODUCTION The Orchid family (Orchidaceae) represents the most advanced family in the Monocotyledoneae and is mostly adapted to the tropics (Frank 1968, Agnew and Shirley 1994). They comprise about 20,000-2500 species in several hundred genera making it one of the 3 largest families of Angiosperm. All forms of orchids are perennial epiphytes or terrestrial and occupy wide ranges of ecological habitats as well as exhibiting highly specialized morphological, structural and physiological characteristics (Dressler 1990) and most orchids have showy flowers. Some of the terrestrial types have tuberous, bulbous thickened roots that are edible. This category is addressed in this paper. Various genera belonging to the orchid family including, Habenaria, Disa and Satyrium are known to form edible tubers (Davenport and Ndangalasi 2003) and Vanilla whose pods provide spices (Frank 1968). Habenaria, Disa and Satyrium are found in the Southern Tanzania (Cribb and Leedal 1982 Lovett et al. 1995 Temu and Chihongo 1998 Davenport and Ndangalasi 2003). Many economically important species of these genera that are common in Tanzania have yet to be identified. All edible orchids are terrestrial and are characterized by stout stems arising from root tubers of various shapes. They occur in varied shapes, sizes and colours and inhabit Nyomora. – Distribution and abundance of the edible orchids ... 46 spectacular natural landscapes on the undulating hills and valleys that typify the Southern Highlands. Despite the differences, the flowers are typically orchidaceous i.e. having showy flowers with a corolla of 3 petals one of which is a labellum or lip that differs from the others and may be spurred as described by Kurzweil (2000). All are deciduous and the aerial stem dies back at the end of the wet season when the plant become dormant, to sprout again at the beginning of the rains from a single bud/’eye’ at the apex of the storage mother tuber(s) that varies from 1-3. Some orchids of Tanzania provide edible tubers from which a staple food known as ‘Kikande’, ‘Chikanda’ or Kinaka is made that is consumed by the Ngoni, Fipa, Nyiha and related tribes in neighbouring Zambia (Davenport and Ndangalasi 2003). Consumption in Zambia is increasing and now threatens orchid populations in Tanzania (Davenport and Ndangalasi 2003) due to over-harvesting ventures on the Tanzanian side. None of these edible orchids have been included in the mainstream staples of Tanzania. Yet, it is known that these indigenous crops are very important in bridging the seasons of hunger to the next harvesting season and therefore can contribute to food security (Temu and Chihongo 1998). One orchid tuber gives rise to only one plant by the vegetative means of reproduction. This coupled by a lack of information on the extent of seed viability and seedling survival, over-harvesting and the competition with pyrethrum and potato cultivation is endangering orchids. Possible domestication and conservation measures (in-situ and ex-situ) require knowledge of the germplasm available and this represents the focus of this study, which was aimed at determining aspects of the extent of distribution of edible 0rchids in the Southern Highland regions of Tanzania namely Iringa, Mbeya Ruvuma and Rukwa. METHODS Sampling and identification A field survey was undertaken through 3-24 March 2002 in the Southern Regions of Tanzania. Visits were made to the District Forest/Natural Resources offices and selected villages in Njombe, Makete, Mbeya, Ileje, Mbozi and Sumbawanga although one of the routes from Kitulo through Iniho to Makete was not passable due to heavy rains (Figure 1). A structured questionnaire was used to interview key informants including district and village council representatives and key players in the trade (harvesters, traders, consumers), followed by on site visits a c c o m p a n i e d b y t h e orchid collectors/harvesters to authenticate dialogues. A total of 50 orchid stakeholders were interviewed including Natural Resource Officers, harvesters , a n d herdpersons/collectors. At each site, 3 sets of plant samples were collected for further identification. Such samples were carefully uprooted, prepared into specimen vouchers, dried and transported to the University of Dar es Salaam (UDSM) herbarium. One set was used for identification at UDSM herbarium using appropriate keys and existing comparison vouchers, one set was donated to through Ben Bytebier for names confirmation while the other was pressed and kept for future referencing. Tanz. J. Sci. Vol 31(1) 2005 47 Figure 1: Map of the Southern Highlands of Tanzania (from Davenport and Ndangalasi, 2003) Distribution and Abundance At each representative site, 3 quadrants measuring 3m x 5m were randomly made along a 250m transect that took the landscape of the site into account. Orchid types and their numbers in each of the 3 quadrants were counted and recorded (Feinsinger 2002 Sutherland 2002). The resultant compiled data were used to calculate the species Diversity Index expressed as: H’ = -_ Pi Ln Pi Where H’ = diversity index; Pi = proportion of i species found at the sites. The diversity indices from both sites were compared using a statistical test at 5% probability (Zar 1984). Plant species abundance and distribution were analyzed using CANOCO package where the data Matrix (species x sites) was prepared using Microsoft excel spreadsheet to enable CANOCO 3.10 analysis (ter Braak 1987). While using CANOCO package, Principle Component Analysis (PCA) was chosen as a technique to show differences in species abundance among study sites. RESULTS AND DISCUSSION It was necessary that the survey be done in February-March when some of the plants were in flower but it was not the most convenient time for accessibility to some sites. Orchid types As presented in Table 1 and 2, a total of 13 edible orchid species belonging to the genus Habenaria (4 species), Satyrium (4 species) and Disa (5 species) were observed and identified during the ethnobotanical survey. These were Harbenaria cornuta Lindl., H. praestans Rendle, H. adolphii Schltr., H. Humilior Reichb.f., ; Satyrium sacculatum (Rendle) Rolfe, ; S. crassicaule Rendle, S. chlorocorys Rolfe and S. breve Rolfe; Disa walleri Reichb.f., ; D. erubescens Rendle, D. leucostachys Kraenzl., ; D . t a n g a n y i k e n s i s Summarch., a n d D. aequiloba Summarh. This may represent only about 15% of the edible orchid species found in the area (Davenport and Ndangangalasi 2003). That all the orchid species found belonged to genus Habenaria, Satyrium and Disa is in agreement with the Nyomora. – Distribution and abundance of the edible orchids ... 48 work of Leedal (1975), Cribb and Leedal (1982) whose general emphasis on all orchid species speculated as many as 200 species as well as Davenport and Ndangalasi (2003) whose study on the orchid trade in the Southern Highlands showed that over 4 million tubers are harvested each year and that as many as 85 species from these three genera are being harvested. Also found were 2 non-edible orchid species later identified as Pteroglossaspis eustachya Reichb.f. and Holothrix longiflora Rolfe. No vernacular names were given by the interviewee for these non-edible orchid species. Table 1: Orchid species identified at visited locations in the Southern Highlands