The current status of knowledge of herbal medicine and medicinal plants in Fiche, Ethiopia

Background A majority of Ethiopians rely on traditional medicine as their primary form of health care, yet they are in danger of losing both their knowledge and the plants they have used as medicines for millennia. This study, conducted in the rural town of Fiche in Ethiopia, was undertaken with the support of Southern Cross University (SCU) Australia, Addis Ababa University (AAU) Ethiopia, and the Ethiopian Institute of Biodiversity (EIB), Ethiopia. The aim of this study, which included an ethnobotanical survey, was to explore the maintenance of tradition in the passing on of knowledge, the current level of knowledge about medicinal herbs and whether there is awareness and concern about the potential loss of both herbal knowledge and access to traditional medicinal plants. Methods This study was conducted using an oral history framework with focus groups, unstructured and semi-structured interviews, field-walk/discussion sessions, and a market survey. Fifteen people were selected via purposeful and snowball sampling. Analysis was undertaken using a grounded theory methodology. Results Fourteen lay community members and one professional herbalist provided information about 73 medicinal plants used locally. An ethnobotanical survey was performed and voucher specimens of 53 of the plants, representing 33 families, were collected and deposited at the EIB Herbarium. The community members are knowledgeable about recognition of medicinal plants and their usage to treat common ailments, and they continue to use herbs to treat sickness as they have in the past. A willingness to share knowledge was demonstrated by both the professional herbalist and lay informants. Participants are aware of the threat to the continued existence of the plants and the knowledge about their use, and showed willingness to take steps to address the situation. Conclusion There is urgent need to document the valuable knowledge of medicinal herbs in Ethiopia. Ethnobotanical studies are imperative, and concomitant sustainable programmes that support the sustainability of herbal medicine traditions may be considered as a way to collect and disseminate information thereby supporting communities in their efforts to maintain their heritage. This study contributes to the documentation of the status of current traditional herbal knowledge in Ethiopia.


Background
Ethiopia has been described as one of the most unusual and important sources of biodiversity in the world [1], yet is perilously close to losing much of this rich diversity due to deforestation, land degradation, lack of documentation of species in some areas as well as of traditional cultural knowledge, and potential acculturation [2][3][4][5]. Intertwined with the irretrievable loss of important species of animals and plants is the risk of loss of traditional herbal medicine knowledge.
An estimated 80 to 90 per cent of Ethiopians use herbal medicine as a primary form of health care [6][7][8][9]. Despite significant recent improvements in modern health care, many rural communities continue to have limited access to modern health care due to availability and affordability [10,11]. It is widely acknowledged that the wisdom of both professional and lay healers in applying traditional medicine to support health and manage illness may be lost to future generations unless urgent efforts are made to document and disseminate the knowledge [3,4,7,12,13] and to engage the younger generation who may no longer be interested in learning the traditional methods [4,7,14]. Therefore Ethiopians, particularly those in rural areas, face an uncertain future in regard to ready access to affordable modern medical services and access to their traditional remedies.

Tradition
Herbalism is one aspect of traditional medicine practice in Ethiopia as it is in many other countries [15]. Herbs have traditionally been used in the home to treat family sickness, and occasionally traditional healers may be consulted. Traditional healers may be from the religious traditions of Cushitic Medicine, regional Arabic-Islamic medical system, or the Semitic Coptic medical system practiced by Orthodox Christian traditional healers [3], who are also referred to in Amharic as debteras. There may be many variations in approach within each system [16]. Spiritual methods are often used in combination with herbal applications particularly by the debteras, and the knowledge is traditionally passed down through the male line. When it comes to household herbal knowledge in the lay sphere, it is also generally considered that knowledge, in accordance with tradition, is preferentially passed on to a favourite child, usually a son [3,12,17,18], although a 2003 study by Gedif and Hahn [17] into the use of herbs for self-care, which primarily interviewed mothers, acknowledged mothers as the "de facto healers of the family treating accidents and ailments with medicinal plants".

Significance of the study
This study examined whether (i) knowledge was transferred to the current generation of lay community members in Fiche, (ii) lay people are knowledgeable about the medicinal use of herbs, (iii) lay people continue to practice herbal medicine in the treatment of sickness within the home. An aim of the study was also to determine whether or not there is enthusiasm for the preservation of knowledge and skills for future generations. The ethnobotanical survey that constituted part of this research helped to identify the plants used by local community members, for future planting in their household and community gardens. To our knowledge, no ethnobotanical exploration had previously been conducted in this area (personal communication, TA). The information gained from this study may inform further studies and projects aimed at documenting herbal knowledge in communities and supporting continued practice and sustainability of traditional herbal medicine in Ethiopia and elsewhere.

Materials and methods
This case study was conducted using an oral history method, a technique for historical documentation which mirrors the cultural practice of passing on knowledge as an oral tradition, and encourages the subjects to present their experience of a specific event or period in the past [19]. It is a process of narrative building and within that framework the story of domestic life emerges. This gives contextual background to the information. A thematic analysis was applied to all interviews.

Ethics
Official collaboration with, and permission from, the Ethiopian Institute of Biodiversity and Addis Ababa University to conduct research ensured that the collection of local medicinal knowledge was compliant with current Ethiopian regulations relating to Access and Benefit Sharing. Ethics approval (No. ECN-10-24) from the Human research Ethics Committee of Southern Cross University was granted, and verbal permission was sought from and granted by each informant, with full explanation given in the local language as to the purpose of the research. Permissions were recorded on film.

Participants
The focus of the case study was the town of Fiche, in the North Shewa Zone of Oromia Region, Ethiopia. Fiche is located 115 km north of Addis Ababa, 9°48′N and 38°44′E, at an elevation of 2700 metres above sea level, with a town population in 2007 of 27,493 [20] ( Figure 1).
Fieldwork was conducted in January and February 2011. Six informants were initially recruited via purposeful sampling by a tertiary-educated, local representative who is knowledgeable about local herbs (referred to herein as 'M8') and who is planning a herbal garden at Fiche (called "Doyu-Armon"). M8 speaks English and provided some translation. The criterion for the sampling was being known in the community to have knowledge of medicinal plants and their use to treat ailments. Further informants were recruited thereafter by snowball sampling. The 15 informants consisted of 14 community members (8 males and 6 females) and a professional herbalist (male) of the Ethiopian Orthodox Christian tradition. In addition to the professional herbalist, three of the males and two of the females were considered by the community to be particularly skilled in herbal knowledge. Informants were aged between 39 and 70, with an average age of mid-forties. Informants are referred to as Male (M) or Female (F) and assigned a number.
Informants' education levels varied from illiterate (80% of informants), to secondary school education completed (10% of informants), with one tertiary-educated informant (M8, who initiated the recruitment of informants and provided some translation) and they belonged to either the Amhara or Oromo ethnic groups. All spoke Amharic and one (M8) was also fluent in English. In addition to the informants, some incidental data was contributed by one of the authors (TA of the Ethiopian Institute of Biodiversity) in his capacity as translator and collector of voucher specimens.
The first informants recruited (2 women and 4 men including the professional herbalist) were identified by the local representative (M8) as persons with significant relevant knowledge, and subsequent informants were recruited by snowball sampling. This sampling method was effective and convenient as it utilised local knowledge to identify appropriate informants.
The first focus group (FG1, six people) provided an introduction of the lead researcher to the community and established the reasons for her presence. Following this session, more people came forward, interested in being part of the process. The professional herbalist was considered a respected Elder and his encouragement to the group was evident. The field-walk/discussion sessions were conducted in two household gardens and the escarpment (open pasture) above the River Jemma Gorge. The market survey was conducted at the Saturday market in Fiche, and the information was obtained from the vendors of the herbs who were mainly women.

Data collection
Field data were collected on six days during January and February 2011. A combination of focus groups (3), individual interviews (5), field-walk/discussion sessions (4) and one local market survey were conducted, with a tertiary-educated translator present at each session. Interview sites, all of which were in Fiche, were: Household garden (HG), homes of community members (H1 and H2), Doyu-Armon garden site (site for planned garden) (D-A), Escarpment above River Jemma Gorge (E) and Fiche Saturday market (M). The Jemma River is a tributary of the Blue Nile. Table 1 shows the timetable of fieldwork.
Additional file 1 shows a plant collection site on the escarpment above River Jemma, as well as extracts of interviews.
The plant specimens collected by the Ethnobotanist (author TA) with the assistance of the informants were pressed, dried and identified following standard procedure and lodged at the EIB Herbarium in Addis Ababa. Translation was provided by TA and M8. All interview and focus group session translations were transcribed directly onto computer by the lead researcher, and all sessions were filmed, with the permission of participants. Later viewing of film footage provided useful review of data. In this way visual dynamics between informants could be viewed and further nuance from discussion picked up without the distraction of the recording process. Footage of 2 focus groups was viewed by a second translator to check areas where translation was indistinct, ambivalent, or not understood by the principal researcher. Other discussions, researcher observations and comments were recorded by hand into a notebook at the time, and a daily journal of all activities, with observations, comments and reflections, was written at the end of each day.
Interviews and focus groups were semi-structured. In an effort to ensure the women and men contributed equally during the mixed focus group discussions, an opening question ("How did you learn?") was directed to each person individually. In this way, informants were able to provide in-depth answers in an individual manner as well as collectively. Occasional prompting, especially on the field-walk activities, would include the questions "What do you use this herb for?" How do you use this herb?" and "What do you call this herb?" allowing uninterrupted flow of discussion unless it strayed significantly from the topic, in which case an appropriate question was asked. Some contextual information was given by the free discussion in this way, often providing additional (unprompted) cultural background.

Data analysis
Grounded theory was applied as a method to conceptualise the data and identify themes. Grounded theory is a method which allows themes to emerge through analysis of data and may provide further deep, thick context to a theory by exposing underlying processes [21]. In keeping with this approach to interpretive analysis, transcripts from each interview were analysed repeatedly to identify emerging themes, and concept codes were assigned (open coding). Coding formed the basis for categories, and the data were examined within categories. Seven category headings were identified and under these all the data were accounted for. Data were examined for herb names, for disease names, and for formulas or prescriptions, and a quantitative list constructed The existing literature was examined for documented uses in Ethiopia of the herbs mentioned and included in this list as a commentary.

Results and discussion
Given that the research was conducted in a language and culture different from that of the principal researcher, some discussion of method with this aspect in mind is pertinent.
The intensive biography interview style of data collection associated with the oral history method allows a researcher to learn about informants' lives from their own perspective [22]. The open discussion of memories, within the context of talking about herbs given to an informant as a child, gave the researcher the opportunity to observe and learn about informants within the context of their home life. Traditional medicine studies undertaken in Ethiopia are not often conducted in this way, with the perspective of an outsider exploring the current situation of the threat of loss of an important tradition, keeping cultural context at the forefront. Whilst being an outsider may on the one hand be seen as a limitation, on the other hand the researcher's presence and interest in their plight highlighted outside interest and gave the community a sense that others considered their knowledge important and of value. The potentially negative issue of being an 'outsider' was ameliorated by the facts that the principal researcher is a herbalist in her own country, is able to speak a little of the language, was introduced to the community by a trusted member of that , also demonstrated tangible ongoing support to the community beyond the research programme. According to Bryman [19], oral history testimonies have provided a method for the voices of the marginalised to be heard. It is not just people who may be marginalised, but also cultural traditions. In respect to the community group in Fiche, important cultural traditions and associated knowledge may be marginalised because community members may not have a strong voice in determining the future of those traditions. Further, the female members of this community may find their knowledge marginalised because despite the acknowledgement that women practice herbal medicine in the home [17,23], the prevalent belief [3,12,17,18] is that men (both professional traditional healers and in the family) are the prime holders of the knowledge. Time constraints of daily household chores may further restrict women's participation in both receiving and passing on knowledge, and having that knowledge may not receive the importance it deserves [9].
The grounded theory approach to analysis was helpful, especially given the particular complexities associated with this study viz. the principal researcher was collecting data while immersed in a language, culture and environment different from her own. Repetition of certain words (translated) provided an opportunity to identify themes. For instance, the word "learnt" appeared at least once per person interviewed in describing different events, not surprising given the question asked but this provided a focus for analysis on first pass. In association with the words "learnt" or "remembered" would be a reference to a family member or influential person. The word "childhood" appeared frequently in this context. Another theme that emerged related to accessibility, availability and sustainability of herbs with subcodes referring to "disappeared", "inaccessible", "not available", "hard to find". Once emergent themes were identified, data were fragmented to lift coded elements out of the context of each interview [24] to list comments and information by group. Fragmented data were then reconnected and reviewed within the context of each interview. Throughout data collection, the researcher was critically aware that words emerged via translation and might have been influenced by translator bias. Mindful of this, the researcher would at times repeat the answer and ask for it to be translated back to the informants for verification. Table 2 lists the themes that emerged from coding.
Fourteen lay community members (6 females and 8 males) and one professional herbalist provided information about 73 medicinal plants from 42 families. Voucher specimens of 53 of these, representing 33 families, were collected and deposited at the Herbarium of the EIB in Addis Ababa. The families contributing the most taxa were Asteraceae (6), Solanaceae (6), Lamiaceae (5) and Fabaceae (5). The major classes of indications cited by informants were gastrointestinal complaints (25 plants) including megagna (12), tapeworm infection (8) and hepatitis (5); psychiatric conditions (7) and respiratory complaints (5).
All herbs named, their uses, and a comparison with uses elsewhere in the literature, are shown in Table 3.
Each informant contributed information about the herbs with which they were particularly familiar. Because discussions were allowed to flow in an unstructured way, this did not lead to a fidelity rating for all the herbs as agreement was not specifically sought from each informant on any one herb and no prompts were given. The two occasions where there was significant consensus on  The herbs are dried, chopped together and put in a cotton pouch to be hung around the pregnant woman's neck in the seventh month. When the baby is born it is taken off the mother and put on the baby M3 "To be collected on a Wednesday or a Friday, having abstained from sexual relations, and having not spoken to anybody on the morning of the collecting day. The herbs are dried outside the house, chopped together and put in a cotton pouch. The cotton must be spun by a lady in menopause, and spun with her left hand not her right hand. The pouch is put on the lady's neck and as soon as she gives birth it is taken from her and put on the baby's neck….this is my specialty" Anti-fertility [26] Fresh pulverised leaf or its juice is placed in the nostril or its juice is sniffed for epistaxis. The crushed fresh leaf is also placed in the genitalia as a remedy for menorrhagia and to stop postpartum haemorrhage [27] Herpes zoster, blood clotting [28] Wounds (kusil) Leaves rubbed and put on cut or wound

F1
Wound [29] Wound [30] Vaginal fumigation [31] Acokanthera schimperi (A.DC.) Schweinf.    Take the dried skin of a hyena and put the herbs in a pouch of the leather as a charm around the neck.

M3
Powder of fruit mixed with water and taken as taenicide [52] With Musena (Albizia anthelmintica) and Nug (Guizotia abyssinica), taken with a drink of Tella (local beer)

M1
"Must be taken simultaneously with Tella. Drink, then jump up and down to dissolve internally.

M1
"Enkoko and Musena are both deadly" With Meterre (Glinus lotoides) and Kosso (Hagenia abyssinica) M8 "I remember my mother giving me this combination" The ripe fruits are collected and the exocarp removed. Fruit swallowed directly using water F2 M3 "It is ok to take Enkoko, Musena and Nug together"   Indigofera zavattarii Chiov.

M3
Fabaceae Jasminum grandiflorum L.  Aphrodisiac, gastritis, headache, ringworm, buda beshita (evil eye) mich (fever with headache) [12] Stomach ache [30] Leucas abyssinica (Benth.) Briq.  "If the juice of Demakese is red when the herb is rubbed by a person, then the person has mich. If it is green, it is not mich. The mother or the daughter will apply this" Kusil (wound), mich (fever) [28] Mich [29] Mich  "The debtera will write a charm with the filtrate and put it on her neck, and the blood will stop" Prepared by debtera: A potion is prepared, buried in the ground for a week. When opened, the inky fluid is used as an ink to write a spell, or charm. Alternatively, the ink is used to tattoo into the skin with a needle M3 "The debtera will use this with other herbs to make a potion. This is put in a bottle and buried for seven days before September 11 (Addis amet -New Year's Day). When opened it will have an inky constituency. The debtera will then use a pen made from Arundo (bamboo), and will write on white paper. It is then worn on the neck. Another way is to tattoo the ink into the skin with a needle" To keep the wife from straying  Debate on this application. Some say the woman should sleep with her husband on the day she takes the medication. "If she sleeps with her husband the ovary will not be badly affected" (M1). "If she goes to the doctor they will clean up that one and she will become pregnant" (F1). "She has to continue sexual relations to stop her ovary being badly affected" (M1). "She has to go to hospital" (M3). Some say it does not matter; used as a contraceptive, the woman will stay without child for 5-6 years. If she wants to become pregnant, she has to take an antidote (merfchow) another plant. M7 says "If she takes the endod she is permanently sterile". F1 says "If you spray poison on a flower, it will die".     Taverniera abyssinica A. -Rich.

2003
Rhesus factor problem in pregnancy (shotelay) as part of formula-see Achyranthes aspera

F2
Verbena officinalis L. Aqwarach/ 1950 Tonsillitis Chewed F5 "My mother would chew it, and she has to take 2 birr* for this. Unless they take the money they cannot be cured. If you refuse, it does not work" Leaf and/or root juice taken against diarrhoea. Decoction of leaf employed as gargle for tongue disease, sore throat and toothache [70] Attuch/

Verbenaceae
Telenz/ Hulegeb *Birr is the unit of currency in Ethiopia Dysentery, digestive after eating raw meat, eczema, eye disease, heart disease, heart pain, indigestion, induction of diarrhoea and emesis to relieve indigestion, insomnia, liver disease, malaria, mumps, snake/ rabid dog bite, sore throat, stomach ache, stomach trouble, tongue disease, tonsillitis [12] Stomach disorder, Herpes zoster, ear problems, evil eye, snake bite, ascariasis [28] Withania somnifera (L.) Dunal Decoction of the root powder taken for rheumatoid arthritis. Bark powder mixed with butter applied as a remedy for swelling [71] Solanaceae Bad spirits (Satan beshita) Adaptogen: whole system F1 "Gizawa is an all-out treatment for the whole system" Evil eye (buda) F 1 "Gizawa is like salt, it can go with anything. For devil spirit, epilepsy, buda. Not for wounds or physical sickness." Evil spirit exorcism, joint infection, arthritis, malaria [12] Chest pain, mich, typhoid, evil eye [68] M8, PH Old saying: "Why did your child die if you had Gizawa growing in your garden?" Narcotic properties. Decoctions are used as pain killers [46] Main actions: Adaptogen, antioxidant, antibacterial and antifungal, anti-inflammatory, chondroprotective, anticancer, anxiolytic and antidepressant [72]  The eyelid is peeled back and rubbed with the back of the leaf. The eyes should be covered and protected from the light until healed.

F1
Widely used as a medicinal plant worldwide Zingiberaceae Unexplained stomach ache (megagna) As part of formula with Ariti (see Artemisia absinthium) and Tenadam (see Ruta chalepensis)
use of herbs for specific diseases was in the discussion of herbs for taeniasis and the discussion of the use of Calpurnia aurea for childhood diarrhoea (see Safety).

How herbal knowledge was acquired
All of the informants (15) described memories of being treated with herbs for illness as a child. All said they subsequently continued to learn, either from parents or knowledgeable elders, or both (see Table 4) The two males who had learnt from both parents said that they had learned more from their fathers. One male who learnt only from his mother said that his father had died when he was young. The professional herbalist had learned from both his grandfather (a priest) and his mother.

Awareness of loss of herbs
There was recognition that some herbs are becoming less accessible, in part due to land degradation and accessibility. When the professional herbalist raised this issue during focus group 2, there was agreement from all present (6 men and 3 women). Examples of comments are: "In the old days herbs were everywhere around the house and in the backyard because people planted them, and also they were growing naturally (referring to the observation in the past that herbs were tolerated or encouraged to grow around human habitation). Now I have to travel for two days to find some herbs. Even in the forest areas, some don't exist any more at all…Now everyone is looking for herbs, but no-one plants and looks after them" (PH) "There is degradation of land, deforestation. Marginally the herbs are still available" (F4) "Initially the Set eret (Aloe pulcherrima) was found close by, but now it is difficult to find this plant, it is only in inaccessible areas now" (M6)

Conservation of herbs
Informants demonstrated an understanding of conservation practices in their wildcrafting of the herbs. When Aloe pulcherrima plants were dug up during a fieldwalk/discussion session (W4), the underground stems were planted for future growth, and an informant helping with collection and identification said: "We don't want to take the whole plant because we use that to keep it growing here" (M5) In a focus group session (FG1), conservative practices were referred to by the professional herbalist: "Some use six herbs for this [formula]. This means more uprooting of plants. I will use only three herbs for this, that means fewer plants used" (PH)

Passing on knowledge
Following a discussion as to whether the younger generation is less likely to be interested in learning about herbal medicine, some informants underscored this issue with their own family experience: "Of my 29 children, four (male priests) have been taught. Two of the children of the priests are interested, two are not" (PH) "I have five children. If they are interested, I will pass it on" (M3) Community awareness of the threat to the future of traditional herbal medicine has been noted elsewhere in Ethiopia [14].
It has been stated that the younger generation in Ethiopia is increasingly losing interest in learning about the herbs [13,29]. However three children (boys between seven and ten years of age) who joined the field-walk/ discussion activities offered some information about the herbs they saw. A nine-year-old boy who worked as a shepherd at the site of a field-walk/discussion excursion, demonstrated in-depth knowledge including recognition and use of medicinal herbs. He was the son of an informant considered a skilled herbalist. The fact that these boys were children of informants, who were knowledgeable about the herbs and used them medicinally, meant that they were more likely to have been exposed to herbal lore in the family setting.
With the possible exception of some herbal medicine education included in religious instruction (there are some known ancient texts held by the Church), due to *Others learnt from include "relatives", "grandmother", "nun", "people around the church".
illiteracy or lack of time, recipes or formulae for herbal treatments continue to be taught to family members solely by demonstration and practical use in the oral tradition of their antecedents. There is a frequently stated understanding that secrecy is an obstacle to the sharing of knowledge, particularly in the domain of the predominantly male professional herbalists [4,68,74]. In contrast to this, and perhaps reflecting increased awareness of the potential for loss, the professional herbalist at Fiche was keen to be involved and fully supported the Botanica Ethiopia objectives of establishing herbal gardens (Additional file 2), contributing and encouraging discussion and collaboration. When the purpose of the research was explained, he said: "Teruneew. (It is good). This must happen. What we are doing is important for the herbs" Another professional herbalist in the area later supported this statement during a spontaneous conversation. The fact that both herbalists were supportive of the establishment of a community "healing herbs" Association as part of the Botanica Ethiopia initiative, with one of the herbalists becoming Deputy Chairperson of the Association, firmly demonstrated willingness to participate in sharing knowledge.

Safety
All participants showed awareness of safety issues and dosage importance.
The importance of safety was discussed in relation to dosages of herbs used for contraception, for children, and with herbs known to have strong activity against taeniasis (tapeworm infection). A focus group debate (FG3) centred on the use of the herb Phytolacca dodecandra (Endod) for contraceptive purposes.
"I gave this to my wife and she never fell pregnant again. Once you take it you are sterile for life" (M5) "If you spray poison on a flower, it will die" (F1) Discussions of herbs used for taeniasis showed consensus in the use of certain herbs (FG1, FG2 and FG3), but debate arose around safety in combining the herbs (FG2). Taeniasis is an epidemic infection in Ethiopia, largely due to the custom of eating raw meat [75]. The discussions focused on four herbs: Glinus lotoides (Meterre), Embelia schimperi (Enkoko), Albizia anthelmintica (Musena) and Hagenia abyssinica (Kosso) with Guizotia abyssinica (Nug) used as a binder to make a paste with the other herb(s). Informants were concerned about the potential for these herbs to cause toxicity and debated the merits of combining what they described as potent herbs. Each of the informants agreed that the four herbs mentioned were important, but there was disagreement as to whether they should be combined (considered dangerous by some) or used separately, and there were varying opinions on how the herbs should be taken. Table 5 summarises this discussion.
In this context, it is interesting to look at whether there has been exploration of the use of these herbs for taeniasis elsewhere. Animal and in vitro studies have been conducted on Glinus lotoides, Embelia schimperi, Albizia anthelmintica and Hagenia abyssinica. In 2006 a paper demonstrating the safety of Glinus lotoides as a taenicidal herb was published [76] but a subsequent investigation showed potential for genotoxicity in mice [77]. There have been investigations into the toxicity and therapeutic activity of a number of herbs traditionally used for taeniasis, including the herbs mentioned by the group in Fiche: Albizia anthelmintica, Embelia schimperi, Glinus lotoides, Hagenia abyssinica and Myrsine africana [75,[78][79][80]. One of these studies reported Myrsine africana to have 'lethal action against tapeworm' [79]. The repeated mention by the informants of this group of herbs in the context of treatment of tapeworm infection contributes to existing documentation of their traditional usage in Ethiopia [3,17,30,75,77] and warrants further pharmacological investigation for their medicinal value.
Another example of a discussion of herbal safety occurred in a focus group (FG3) and concerned the use of Calpurnia aurea (Digita) for the treatment of childhood diarrhoea. The dosage, strength and potential toxicity of this herb were discussed.
"Take the young shoots from seven plants of Digita, rub the leaves in the hands for juice, for children with diarrhoea (tekmat). Put juice into water depending on the age of he child, dosage is very important. It is very strong. Very small by spoon. One teaspoon. Just once" (F3) "It can be very dangerous. They [informants] say the stem bark is poisonous. Only the young shoots are used and even then one has to be very careful" (M8, also translating) "Actually it can send you crazy. If you go crazy, you will die" (F1) "It should be measured carefully" (F2) The use of Calpurnia aurea, a quinolizidine alkaloidcontaining member of the family Fabaceae, for the treatment of diarrhoea and a range of other conditions, is well documented from Ethiopia and other parts of Africa [28,45]. It has demonstrated anti-diarrhoeal effect in mice and in vitro inhibitory activity against a range of diarrhoea-causing bacteria [45].

Gender
The literature frequently discusses the Ethiopian tradition of preferentially passing on knowledge in the male line, either through the Church tradition or within the family [3,12,17,81] and studies tend to show that men have better medicinal plant knowledge [4]. However in focus group 1 (5 men and 1 woman), when one of the men declared that women hold more knowledge, all agreed that women have more herbal knowledge than men relating to the use of medicinal plants in the home.
"In the countryside, the women hold all the knowledge…the women had to learn the hard way, because men could be away at war or simply not there, so the women left behind have to take care of themselves and their children" (M1, with agreement from PH, M3, M4, M8, F1) "More women know about the application of the herbs" (PH) This concurs with the findings of Fassil in her 2005 study of home-based medicinal plant use in rural communities in the Bahir Dar Zurie Wereda (district) in northwest Ethiopia [9], which showed that women have particular roles in traditional health care delivery in their capacities as mothers and cultivators of home gardens, and also the 2003 study by Gedif and Hahn [17] which recruited mothers as informants.
Group discussions were not so effective at capturing the information of the women as they were often pressed for time and unable to be present for as long as the men. Even during interviews the women were busy with children or food preparation. This limitation was also noted by Fassil [9].

Herbs: identification and usage
Both men and women on the field-walk/discussion activities demonstrated ability to identify medicinal herbs. At the Saturday market, women were the vendors of the herbs and were knowledgeable about their uses. The Saturday market was attended by members of the Fiche community and surrounding towns, with a variety of stalls managed by men, women and children selling foodstuffs (including culinary herbs), household equipment, and medicinal herbs. From a survey taken at the Saturday market, 15 medicinal herbs were identified (Table 6).

Conclusion
This study has shown that herbal medicine continues to be of great importance to this community in Ethiopia as part of their healthcare system, and they are aware that the knowledge and the herbs are at risk of disappearing. Knowledge continues to be passed on via the oral tradition and by application. This community is motivated to help to increase awareness of, and accessibility to, the herbs they use to treat illness in the family home. There were several important aspects noted during this study that future researchers in the area may wish to consider. One recommendation arising from our experience is that women be released from domestic duties for the purpose of interviews and focus groups. This would allow them to contribute their knowledge and experience more fully.
It may fairly be argued that conducting a study where the principal researcher does not share language or cultural background could present significant obstacles, but there were unexpected advantages that arose from this. The researcher's presence demonstrated to the informants an external awareness of, and respect for, the knowledge held by the community, and for their predicament. The fact that the research supported the implementation of a project to establish a medicinal herb garden in the community also contributed to the willingness of the informants to contribute and share their knowledge. Collaboration with Ethiopian authorities (AAU and EIB) was essential for the successful conduct of the research. It was also important and helpful to consult with local authorities. Local government (Kebelle) and City Council representatives provided administrative support for the formation of the Etse-Fewus (Healing Plants) Association subsequent to the fieldwork, and local government subsequently donated land for a community medicinal garden, giving demonstrable government legitimacy to the initiative.
We recognise that all these elements were critically important for the successful conduct of the research and future researchers are encouraged to investigate how they may best support the communities with which they work. In doing so, they will contribute in part to the United Nations Millennium Development Goals [36], primarily those related to reduction of child mortality, improvement of maternal health, combating HIV/AIDS, malaria and other diseases, promoting gender equality and empowering women, and ensuring environmental sustainability. If Ethiopians lose their traditional herbal medicineeither the knowledge, or the plants or both -they will lose the ability to provide herbal treatment for their families. If they are also unable to access conventional medicine either through lack of affordability or availability, as is still the case in many rural areas particularly, they would be in an unenviable situation. Ethnobotanical, ethnomedical and anthropological research must continue in Ethiopia in order to understand the cultural, sociological and practical considerations that inform the wider community at institutional and governmental level. In the future, Ethiopians should be able to take advantage of opportunities to develop the potential of their rich medicinal plant resources via documentation of knowledge of use and pharmacological investigation of medicinal properties of the plants. Integration of traditional herbal medicine with outreach medical services may be a beneficial outcome of supporting further investigations in Ethiopia's medicinal herb lore.
Financial support Blackmores Ltd. and the Blackmore Foundation provided start-up funding for the Botanica Ethiopia project, and thus supported this research. Blackmores also provided an educational grant to Ed'A. Additional funding was obtained from the Australian African Children's Aid Support Association (AACASA) and individual donations. Donors had no involvement in the study design and execution, data analysis and interpretation, or the preparation of the manuscript.