Indigenous Knowledge on Edible Wild Yams (Kumbu) in the Mount Cameroon Region: Towards Domestication for Enhanced Food Security

Abstract

Growing food insecurity can in part be attributed to a lack of diversity in arable crops, with most African countries now focused on the production of a few “green revolution crops”. Indigenous knowledge of traditional food types could hold the key to the genetic diversification of crop production systems. Wild yams are indigenous crops that have been relegated to the background. This study aimed to assess the state of knowledge in, and cultivation of, wild yams collectively called “Kumbu” by the Bakweris of the Mount Cameroon Region. Following reconnaissance surveys, semi-structured questionnaires were administered to 583 interviewees across 41 villages in this region. Data were analysed in the SPSS version 21 statistical package with significance at α = 0.05 where necessary. Results showed that the study population was fairly balanced in terms of gender (SD = 0.534), with males representing 56.8% of the sample. A majority of the interviewees (53.3%) were married, and most had received at least primary education (85.2%). Most (61.6%) of the interviewees do not cultivate Kumbu due to a lack of available seeds (69.3%) and a preference for other yams (30.7%). Of those who cultivate Kumbu (38.4% of the interviewees), a majority (89.6%) have less than five stands of Kumbu. The different names (10) and types (13) of Kumbu could represent linguistic polymorphism, requiring further studies for proper identification. A majority (68.1%) of the interviewees had no idea of the differences between Kumbu types. Agronomic practices, pests, and disease management reported for Kumbu are similar to those of other mainstream yam types. We conclude that the state of knowledge on Kumbu in the Mt Cameroon Region is limited and on the decline. Bringing Kumbu production to the mainstream requires research on molecular taxonomy, propagation techniques, and agronomic practices for better yields.

1. Introduction

Cropping systems in sub-Saharan Africa, which were mainly for subsistence, used to consist of crops grown in a balanced mixture. In addition to this, non-timber forest products were also gathered from the wild. These systems were ecologically sustainable and sufficiently robust to ensure food security as they were adapted to local environmental conditions and cultural practices. With the advent of colonisation by the Western world, these systems were upended by projects that sought to enforce the production of specific crops for the industry [1]. These led to a large-scale establishment of rubber, oil palm, and banana plantations on the rich volcanic soils of the Mount Cameroon Region. This shift not only displaced traditional food systems but also reduced the diversity of crops grown, undermining local food security in Ghana and Cameroon [2]. This is because plantations often monopolise the fertile land, leaving smallholder farmers with limited access to arable land for food crop production. Additionally, the focus on export-oriented cash crops diverts resources away from subsistence farming, further exacerbating food insecurity. When attention began to be paid to food crop production in the post-colonial era, the emphasis had shifted to high-yielding—called green revolution—crops like maize. While these crops were promoted for their productivity, the focus on a narrow range of crops led to the marginalisation of traditional food crops, many of which were better suited to local conditions. Traditional food crops lost significance [3] and in some cases, knowledge of these crops has been almost entirely lost. However, in most countries, production of the few mainstream crops like maize, rice, grain legumes, etc., has not been able to keep pace with demand, and high yield gaps are reported for these crops, driven by factors such as suboptimal farming practices, limited access to inputs, and the impacts of climate change [4]. High dependence on imports of grains and cereals has, therefore, developed but this is not sufficient to ensure food security [5], and this import-dependence in turn is susceptible to the volatilities in the global market and political space. For instance, since the start of the Ukraine war, cereal prices as well as the prices of inputs like fertilisers have skyrocketed, greatly impacting local food prices in Africa and, hence, reducing the gains made in ameliorating food security. This is because Russia and Ukraine were major suppliers of these commodities to many African Countries [6].

The United Nations 2030 Agenda for sustainable development is encapsulated in the Sustainable Development Goals (SDGs), in particular, Goal 2—which aims at zero hunger [7]—as well as Cameroon’s Vision 2035 objectives of a food-safe and food-secure nation [8], which are critical for ensuring equitable access to nutritious food and the eradication of hunger and malnutrition for all populations. This goal and vision are at risk of not being realised if the current trajectory continues. However, the increasing challenges posed by climate change, global market volatility, and the erosion of traditional food systems threaten the achievement of this goal and vision. Therefore, to reduce the impact of international turbulence on local food security, there is a need for food crop diversification, which requires us to re-visit traditional crops that are less susceptible to these international pressures [9].

Root and tuber crops have been a substantial part of Africa’s cuisine since historical times, with the production of yams, cassava, cocoyam, and sweet potato [10]. These crops are not only culturally significant but also highly adaptable to local environmental conditions, making them critical for food security in the face of climate change. Here too, the varieties mostly grown nowadays are those that have been improved and tested over time, with the International Institute of Tropical Agriculture (IITA) at the fore of tuber crop development in general and yam cultivation and improvement techniques in particular. Mainstream yam varieties grown in Africa include Dioscorea alata L., D. rotundata Pior., D. cayenensis Lam., and D. dumetorum (Kunth) Pax [11]. For these, the technology and know-how are readily available. While these varieties benefit from established technologies and know-how, the genetic pool has been significantly narrowed, excluding many wild yam varieties that were once domesticated or harvested from the wild. Apart from having the potential to be more resistant to diseases like anthracnose and yam mosaic virus diseases, wild yam types may also require less inputs for greater yield, and are usually more climate-change-adapted [12]. They also form part of the heritage of African populations, which requires preservation.

Cameroon is one of the most significant yam-producing countries in the world with agriculture being the back bone of its economy. Yam production stands at 714,349.5 tons annually [13]. If Kumbu is valorised and brought into the mainstream, this production of yams could double, potentially doubling the contribution of yams to the nation’s GDP. Also, there would be greater diversity both in terms of the gene pool and consumer choices. With skyrocketing prices for rice and other imported food items, increased tuber crop production including Kumbu would go a long way to improving food security. Indeed, yams are produced in all five agroecological zones of the country, with emphasis on a few mainstream species. There has been a lot of genetic erosion of local yam species in preference for the more developed/commercial species like D. alata and D. rotundata varieties. Traditional yam species that used to be cultivated on a small scale or collected from the wild for their cultural and food significance have been relegated to the background, with knowledge on them eroding rapidly. According to Dumont et al. [14], there are 17 species of wild yams in Cameroon, and 10 of these are edible. Laird et al. [15] explored the intricate relationship between plants and peoples of the Mount Cameroon Region and showed that wild yams constituted an important part of indigenous diets, consistent with earlier findings by Dumont et al. [14] on the Baka pygmies of Cameroon. To valorise these wild genetic resources in a rapidly urbanising space, ethnobotanical studies are a first step, especially in assessing local knowledge of the species. In the Mount Cameroon Region, consumption of “wild” yams or yams in the process of domestication has long been practised, side by side with mainstream D. alata, D. rotundata, and D. cayenensis cultivation. Some knowledge of such wild yams collectively called “Kumbu” in the indigenous language may still reside within the population. This research aimed to assess the state of knowledge on “Kumbu” cultivation among the people of the Mount Cameroon Region of Cameroon as a first step towards adding value and bringing this crop into the mainstream for food crop diversification and food security in the region.

2. Materials and Methods

2.1. Study Site

This study was carried out in the Mount Cameroon Region (Figure 1). The Mount Cameroon Region is located in the Fako and Meme Divisions of the Southwest region of Cameroon and lies between latitudes 4.055° and 4.378° N and longitudes 9.031° and 9.294° E. The area shares boundaries with five subdivisions of the country, namely Buea, Limbe II, Muyuka, Idenau, and Mbonge, and includes 41 villages.

The Mount Cameroon Region has fertile volcanic soils that contribute to the expansion of agricultural activities in the area. There are two seasons that are well-defined with a period of heavy rainfall occurring between June and November; with annual rainfall that ranges from 2000 mm in the east–west to 10,000 mm on the coast; and a dry period extending from November to May with annual temperatures that drop by 0.6 °C for every 100-meter ascent. The relative humidity is 75–58%. The area is known for its exceptional plant diversity and high number of endemic species, with 49 plant taxa that are strictly endemic [16].

2.2. Reconnaissance Survey

Before conducting the survey, a semi-structured questionnaire was created and tested with key informants during the preliminary visits to the study locations to assess its relevance to the current research. Based on the feedback from these informants, the questionnaire was subsequently revised. The process for identification of key informants and the pre-testing of the questionnaire involved close collaboration with the Extension Officers from the Sub-Divisional Delegations of Agriculture and Rural Development in Buea, local community members, and traditional rulers, considering factors such as their age, traditional knowledge of Kumbu production, native status, and duration of residence in the community. They willingly shared their insights on the varieties, cultivation, and uses of Kumbu.

An initial survey was conducted to acquaint ourselves with the study site. During this survey, it was observed that 41 villages surround the Mount Cameroon National Park, which, together with the park, constitute the Mount Cameroon Region. Access to these villages was varied but three routes were predominantly used. Villages on the north, northwestern, and northeastern flanks were accessed through Muyuka. This access circumnavigates the Mount Cameroon National Park, exiting at Bibunde or Idenau in the West Coast. The second access is through the Buea subdivision, passing through Bwitingi, to access villages such as Bova I and Bova II. Other villages such as Bokwoango were accessed through the main boulevard of Buea. It was noticed that agriculture is a major activity in the 41 villages surrounding the Mount Cameroon National Park, and crops such as plantain, banana, yams, and various fruit and leafy vegetables are cultivated. In each village, permission was obtained from the traditional authorities to enable the study to proceed.

2.3. Sample Identification and Selection

Yam farmers were the main targets of this study since they have the highest potential to be knowledgeable of the wild yam type “Kumbu” in which the researchers were interested. Therefore, in each village, the traditional authorities provided field guides who could identify the homes of the different yam farmers. Therefore, the sampling approach was purposive.

2.4. Sampling Design and Sample-Size Determination

The method used in this study was the purposive sampling approach, and the data collection instrument was the semi-structured questionnaires. To obtain the sample size for the administration of questionnaires, Slovin’s formula [17] was used, as outlined below:

$$\mathrm{S}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e} \mathrm{s}\mathrm{i}\mathrm{z}\mathrm{e}\left(\mathrm{n}\right)={\displaystyle \frac{\mathrm{N}}{1+{Ne}^2}}$$

(1)

where

• n = sample size;
• N = population size;
• e = acceptable margin of error margin.

The population size (N) of yam farmers in the Mount Cameroon Region = 1479 [18]. Assuming the acceptable margin of error at a 5% confidence interval, or 0.05:

$$\mathrm{n}={\displaystyle \frac{1479}{1+{1479\left(0.05\right)}^2}} = 315$$

(2)

Using the above formula, we obtained a sample size of 315 yam farmers but the effective sample was 583 interviewees in 41 villages of the Mount Cameroon Region (

Table 1. Villages in the Mount Cameroon area where the yam production survey was carried out alongside the sample sizes.

SN	Village	Number of Interviewees	% of Interviewees	SN	Village	Number of Interviewees	% of Interviewees
1	Etome	7	1.2	22	Kotto	8	1.4
2	Upper Boando	6	1	23	Ekonjo	6	1
3	Batoke	8	1.4	24	Bokwai	11	1.9
4	Bokova	13	2.2	25	Mokunda	23	3.9
5	Bova II	60	10.3	26	Masuma	7	1.2
6	Lower Boando	6	1	27	Muyenge	12	2.1
7	Sanje	8	1.4	28	Wonya Emongo	22	3.8
8	Ewonda	6	1	29	Bova I	20	3.4
9	Woteva	20	3.4	30	Bwassa	15	2.6
10	Bonakanda	30	5.1	31	Bokwoango	26	4.5
11	Bakingili	11	1.9	32	Mapanja	15	2.6
12	Likomba	22	3.8	33	Lio La Buea	7	1.2
13	Wonya Lyonga	22	3.8	34	Bwitingi	10	1.7
14	Koke Bwiteva	10	1.7	35	Yasingi	30	5.1
15	Njonje	6	1	36	Bokosso	4	0.7
16	Bebunde	10	1.7	37	Mundame	8	1.4
17	Likoko Membea	18	3.1	38	Ebie	3	0.5
18	Komboni	25	4.3	39	Kuke Kumbu	8	1.4
19	Wondogo	28	4.8	40	Ekona Lelu	10	1.7
20	Efolofo	9	1.5	41	Mundongo	2	0.3
21	Bomana	11	1.9
Total				41		583	100

). This was mainly due to greater accessibility, the assistance provided by the field guides in each village who could exhaust the yam farmers’ population in some instances, and the willingness of interviewees affiliated with NGOs and CIGs to participate.

2.5. Demographic Characteristics of the Interviewees

The 583 interviewees were aged from less than 30 years old to more than 60 years old. No single age group constituted a majority, although more interviewees were aged >60 years (23.7%) followed by the age group 31–40 years. The least-represented age group was those aged 41–50 years (15.1%). The standard deviation (SD = ±1.443) was small, indicating a uniform distribution of interviewees in the different age groups.

In terms of gender, this was fairly uniformly distributed as well (SD = 0.534), with males representing 56.8% of the sample. A majority of the interviewees (53.3%) were married, 30.9% were single, and both widows and divorcees were represented in the sample. From those without any formal education (14.8%) to interviewees who have received a university education (17.7%), the interviewees represented a broad spectrum of education levels (SD = ±0.948), with a majority having received at least primary education (85.2%).

2.6. Data Collection

Semi-structured questionnaires were administered in person to 583 yam farmers in 41 villages of the Mount Cameroon Region. In terms of area, the total coverage of the 41 regions was the target, giving an area coverage of 100%. In terms of the sample size, the effective sample (583) exceeded the expected sample size (315) by 85%. The questionnaire was piloted among a select sample (5) in Bokova, and the participants’ responsiveness to the questions provided 100% of the information envisaged. The questionnaire administration thus proceeded, and administration was conducted in person. In line with the ethical requirements for research at the University of Buea, each participant was read the Informed Consent Form, and we only proceeded after they accepted to willingly participate in the research. There were no minors or dependents in the sample requiring parental consent. All questions were asked in the “pidgin” language, which is a global medium of communication irrespective of the level of education or region of the country. When the interviewees provided information such as the names of Kumbu in the local Bakweri language, these were spelt out by the researcher as pronounced using the English alphabet. Since the questionnaires were administered in person, the response rate was 100%.

A focus group discussion was conducted in each community to document knowledge on Kumbu production, while also validating individual interview responses. Participants brought Kumbu samples for demonstration, enhancing the discussion with practical examples of the types of Kumbu available in their environment. Participants for the focus group were chosen from various socio-demographic backgrounds, guided by extension officers and community elders, focusing on their knowledge of Kumbu production. Discussions followed semi-structured guidelines and interview questions. These discussions aided the research in that they gave insights for better interpretation of the analysed questionnaires.

2.7. Data Analysis

Data were captured in the SPSS statistical package version 21 (IBM Corporation, Armonk, NY, USA) and coded. Where necessary, binary coding was used. Descriptive statistical analyses were conducted. The Chi-Square test of association, Cramer’s V test of association, and Cohen’s Kappa statistic of the variable agreement were used where there was a need to test associations between variables. For significance, α was considered at 0.05. Tables and charts were constructed for the presentation of the results.

3. Results

3.1. State of Knowledge of “Kumbu” Among the Sample in the Study Site

Analysis of the state of knowledge of “Kumbu” among the interviewees shows that a majority (69.3%) of the interviewees indicated that they know what Kumbu is, while a significant proportion (30.7%) indicated that they do not know what Kumbu is. Similarly, a majority (68.5%) of the interviewees indicated that they have seen Kumbu, while 31.5% indicated they have never seen Kumbu.

Figure 2 presents the locations where Kumbu is reported to have been seen by interviewees. These locations include farms, markets, and friends’ residences. A majority (73.51%) of the interviewees indicated that they have seen Kumbu on the farms, while very few (9.16%) indicated that they have seen it in markets.

3.2. State of Cultivation of Kumbu in the Mount Cameroon Region

A thematic analysis of the state of cultivation of Kumbu within the study site is presented in

Table 2. Thematic analysis of the state of cultivation of Kumbu within the study site.

Theme	Frequency	Valid Percent	Cumulative Percent	SD
Do you plant “Kumbu”
Yes	223	38.4	38.4	0.487
No	357	61.6	100
No response	3
Total	583
If “NO” above, why?
No available seeds	217	69.3	69.3	0.462
Prefer other yams	96	30.7	100
Answered “yes” above plus no response	270
Total	583
If you answered “Yes” above, how many “Kumbu” stands do you have?
Less than 5 stands	194	86.6	86.6	0.391
More than 10 stands	27	12.1	98.7
More than 100 stands	3	1.3	100
Total	224	100
Answered “no” above plus no response	359
Total	583
If less than five stands why?
No available seeds	159	71	71	0.867
Difficult to plant	13	5.8	76.8
Longer maturity than other yams	52	23.2	100
Other response to no. of stands plus no response	359
Total	583

. A majority (61.6%) of the interviewees do not cultivate Kumbu. A significant proportion (38.4%), representing 223 interviewees, cultivate Kumbu. Of those who do not cultivate Kumbu, the reasons provided include a lack of available seeds (69.3%) as the most frequent problem, although a preference for other yams (30.7%) was also mentioned. Of those who cultivate Kumbu (38.4%), a majority (89.6%) have less than 5 stands of Kumbu in their farms, with very few (1.3%) having more than 100 stands of Kumbu. Among the interviewees with less than five stands, the reasons provided for this low level of cultivation include a lack of seeds (71%), long juvenile periods compared to other yams (23.2%), and difficulty in planting (5.8%).

3.3. Indigenous Knowledge of Kumbu in the Mount Cameroon Region

Results on the interviewees’ knowledge of the different names for Kumbu within the study site are presented in

Table 3. Alternative names of Kumbu within the study site.

Alternative Names of Kumbu	Frequency	Valid Percent	Cumulative Percent	SD
No idea	538	92.6	92.6	1.721
Water yam	8	1.4	94
Etoh	3	0.5	94.5
Nyake	12	2.1	96.6
Tafftuf	1	0.2	96.8
Ekale	1	0.2	97
Jasse	2	0.3	97.3
Sweet yam	4	0.7	98
Bakweri yam	8	1.4	99.4
Calabar yam	3	0.4	99.8
Yesi	1	0.2	100
No response	2
Total	583

Valid percentage calculated based on effective responses.

. Ten different names are presented. A large majority of interviewees (92.6%) had no knowledge of any alternative names for Kumbu. Of the remainder, most (2.1%) refer to Kumbu as Nyake, followed by Water yam or Bakweri yam (both 1.4%). Other names include Sweet yam (0.7%), Etoh (0.5%), and Calabar yam (0.4%).

There are different types of Kumbu within the study site (

Table 4. “Types” of Kumbu known within the study site.

Types	Frequency	Valid Percent	Cumulative Percent	SD
Enyanke	21	3.6	3.6	1.882
Mwende	19	3.3	6.9
Nyake and Mwende	157	27	33.9
Etoh	5	0.9	34.8
None	358	61.6	96.4
Etale	1	0.2	96.6
Jaseh	3	0.5	97.1
White and yellow sweet yam	4	0.7	97.8
Jeme	1	0.2	97.9
Molemba	1	0.2	98.1
Yono and Kumba Ngenge	2	0.3	98.5
Lulu	1	0.2	98.6
Movoh mo Njoku and Mwah mata	2	0.3	99
Etana and calabar yam	2	0.3	99.3
Ewombo mbwende, yabole and queti	3	0.5	99.8
Etanake and mweni	1	0.2	100
Missing	2
Total	583

Indigenous names are spelt as pronounced using the English alphabet.

). However, a majority of interviewees (61.6%) are not aware of this. Of those who have this indigenous knowledge, thirteen different types of Kumbu are reported. Among these, Nyake and Mwende (27%) constitute the largest proportion. Others include Etoh (0.9%), Yono and Kumba Ngenge (0.3%), and Jase (0.5%). Yam “types” is used as a descriptive term to distinguish Kumbu based on indigenous understanding of their differences. It does not refer to species or varieties stricto sensu as the indigenes may not have this knowledge. There is ongoing research to botanically characterise these Kumbu “types” in order to resolve their classification.

Table 5. Differences between the different types of Kumbu identified by the community.

Differences Identified	Frequency	Percent	Cumulative Percent	SD
Nyake has a bitter taste with colourful tubers	13	2.2	2.2	0.796
Mwende is not bitter and has white tubers	6	1	3.3
Nyake is colourful and mwende has white tubers	107	18.4	21.8
No idea	397	68.1	90.3
Mwende grows straight into the soil like normal yam while Nyake is big and grows flat	45	7.7	98.1
Nyake is rough while mwende is smooth	7	1.2	99.3
Jeme has flat tubers	1	0.2	99.5
Ewombo is big in size, mbwende grows upwards, yobole is yellow, while queti is small in size and dug in the dry season	3	0.5	100
No response	4	0.7
Total	583	100

presents the perceived differences between the different types of Kumbu in the study site. A majority (68.1%) of the interviewees had no idea of the differences between Kumbu types. Of those who could identify some differences, 18.4% of interviewees reported that the main difference between Nyake and Mwende is the colour. Nyake is colourful, while Mwende has white tubers. Another group of interviewees (1.2%) reported that Nyake has a rough tuber surface while Mwende has a smooth tuber surface. Some interviewees (7.7%) say Nyake has big tubers and grows flat (horizontal). Nyake is also reported by 2.2% of the interviewees to have a bitter taste.

3.4. Preference of Different Types of Kumbu Among Interviewees in the Mount Cameroon Region

Two Kumbu types, namely Nyake and Mwende, are the most preferred by the interviewees. Many interviewees (18.01%) prefer Nyake to the other Kumbu types, followed by 8.58% who prefer Mwende. Also, some interviewees (8.32%) prefer both Nyake and Mwende. However, a large majority (62.78%) could not say whether or not they preferred any Kumbu type.

The cross-tabulated results detailing the reasons for interviewees’ preferences of different Kumbu types are summarised in

Table 6. Chi-Square and Cramer’s V tests of association between the preferred Kumbu types and the reasons for preference.

Preferred Kumbu		Reason for Preference				Total	χ2	Cramer’s V
		Good Taste	Yam Colour	Insect Resistance	Others
Nyake	Counts	64	2	3	35	104	χ2 = 36.967, p = 0.017	V = 0.235, p = 0.017
	%	61.5	1.9	2.9	33.7	100
Mwende	Counts	36	9	0	5	50
	%	72	18	0	10	100
Nyake and Mwende	Counts	35	3	0	10	48
	%	72.9	6.3	0	20.8	100
Etoh	Counts	0	0	0	2	2
	%	0	0	0	100	100
No idea	Counts	8	0	0	1	9
	%	88.9	0	0	11.1	100
Movoh no Njoku	Counts	3	0	0	1	4
	%	75	0	0	25	100
White yam	Counts	4	0	0	0	4
	%	100	0	0	0	100
Ewombo	Counts	2	0	0	1	3
	%	66.7	0	0	33.3	100
Total	Counts	152	14	3	55	224
	%	67.9	6.3	1.3	24.6	100

Cramer’s V value ranges from 0 (complete independence) to 1 (complete association). Indigenous names are written in italics.

. The variety Nyake is predominantly favoured for its superior taste, with 61.5% of respondents citing this as their primary reason for preference. Similarly, 72% of those who prefer Mwende also highlighted taste as their main consideration, a pattern echoed by 72.9% of interviewees who expressed a preference for both Nyake and Mwende. A significant majority of participants indicated a preference for Movo no Njoku (75%), white yam (100%), and Ewombo (66.7%), attributing their choices primarily to the appealing taste of these Kumbu varieties. The Chi-Square (χ² = 36.967, p = 0.017) and Cramer’s V (V = 0.235, p = 0.017) tests of association show a significant association between the preferred Kumbu types and the reasons for this preference.

3.5. State of Kumbu Cultivation in the Mount Cameroon Region

A thematic analysis of the state of Kumbu cultivation in the study site is presented in

Table 7. Thematic analysis of state of Kumbu cultivation.

Theme	Options	Frequency	Valid Percent	Cumulative Percent	SD
On your farm, did you plant “Kumbu” or did you find it growing in the wild
	Planted	144	53.9	53.9	0.499
	Found in the wild	123	46.1	100
	Missing	316
	Total	583
How did you obtain the seeds that you used for planting “Kumbu”
	Local/traditional Kumbu seeds	144	57.1	57.1	2.366
	Improved Kumbu seeds	3	1.2	58.3
	Bought from the local market	14	5.6	63.9
	Exchanged with neighbours	18	7.1	71
	Gifted from neighbours	61	24.2	95.2
	Others	12	4.8	100
	Missing	331
	Total	583
If you use local/traditional “Kumbu” seeds, give the reasons.
	Available seeds	174	68.8	68.8	0.748
	Lack of improved seeds	40	15.8	84.6
	Longer viability of seeds	39	15.4	100
	Missing	330
	Total	583
How do you handle your “Kumbu” seeds for your next crop?
	Store on the farm	216	86	86	0.54
	Exchange with neighbours	25	10	96
	Sell to neighbours	10	4	100
	Missing	332
	Total	583
When selecting “Kumbu” seeds for the next crop, what size do you use?
	50 g	7	1.2	1.2	1.093
	100 g	14	2.4	3.6
	200 g	7	1.2	4.8
	300 g	1	0.2	5
	500 g	1	0.2	5.1
	Others	10	1.7	6.9
	No idea	543	93.1	100
	Total	583	100

. A majority of interviewees who are familiar with Kumbu cultivate it on their farms (53.9%), while the rest (46.1%) domesticate volunteer stands (stands maintained from the wild). Of those who cultivate Kumbu, the seeds are mainly obtained from local/traditional sources (57.1%) or as gifts from neighbours (24.2%). The reason most cultivators use traditional seeds is because these are readily available (68.8%). Other reasons include the lack of improved seeds and the greater viability of traditional seeds. In preparation for the next season, a majority (86%) of the interviewees store seeds on the farm from the harvest, while 10% obtain Kumbu seeds through exchanges with neighbours. With respect to the size of seeds for planting, there is a general lack of knowledge among interviewees.

With respect to the level of experience in cultivating Kumbu, 62% of the interviewees had no response. Of the remaining interviewees, 14.54% had more than 15 years of experience in cultivating Kumbu. This is followed by 7.5% of interviewees with 1 to 4 years of experience. The standard deviation of these responses was 2.366, suggesting high variability in the data.

3.6. State of Indigenous Knowledge of Agronomic Practices in Kumbu Cultivation

With respect to the planting distance for Kumbu, a majority (86.92%) of the interviewees indicated that they know the planting distance of Kumbu, while the rest (13.08%) said they do not know the planting distance. Cross-tabulated results of the interviewees’ perspectives on planting distance and their reported knowledge of this distance are presented in

Table 8. Cross-tabulation of interviewees’ perception of knowledge on the planting distance of Kumbu.

Knowledge of the Planting Distance of “Kumbu”		Proposed Planting Distance (m)							Chi-Square	Cramer’s V
		0.25	0.5	0.75	1	>1	No Idea	Total
Yes	Count	3	1	8	59	4	1	76	χ2 = 14.221, p = 0.014	V = 0.144, p = 0.014
	%	3.9	1.3	10.5	77.6	5.3	1.3	100
No	Count	0	0	2	4	0	2	8
	%	0	0	25	50	0	25	100
Total	Count	3	1	10	63	4	3	84
	%	3.6	1.2	11.9	75	4.8	3.6	100

. A majority (77.6%) of the interviewees who profess knowledge of the planting distance reported that Kumbu is planted 1 m by 1 m in the field. This category is followed by 10.5% of interviewees who reported that Kumbu is planted at 0.75 m × 0.75 m. The Chi-Square test of association (χ² = 14.221, p = 0.014) and Cramer’s V (V = 0.144, p = 0.014) show a significant association between the variables.

The state of knowledge on agronomic practices related to Kumbu production assessed from the perspective of interviewees who cultivate the crop is presented in

Table 9. Cross-tabulation of interviewees who plant Kumbu with agronomic practices for the crop.

Interviewees who Plant Kumbu	Plant Density/ha					Chi-Square	Cramer’s V	Cohen’s Kappa
	5000	7500	10,000	15,000	Total
Count	11	14	26	3	54	0.820, p = 0.936	V = 0.112, p = 0.936	0.006, p = 0.874
%	20.4	25.9	48.1	5.6	100
	Staking materials				Total
	None	Bamboo poles	Existing trees
Count	4	26	187		217	13.289, p = 0.004	V = 0.230, p = 0.004	0.027, p = 0.002
%	1.8	12	86.2		100
	Reason for use of selected staking material				Total
	Availability	Affordability	Easy transport	Others
Count	187	24	1	5	217	5.037, p = 0.169	V = 0.142, p = 0.169	0.103, p = 0.073
%	86.2	11.1	0.5	2.3	100
	Use of fertilisers in production		Total
	Yes	No
Count	13	210	223			14.520, p = 0.000	V = 0.160, p = 0.000	0.063, p = 0.000
%	5.8	94.2	100
	Type of fertiliser
	Organic	Chemical	Both	Total
Count	11	1	2	14		0.768, p = 0.681	V = 0.213, p = 0.681	0.019, p = 0.896
%	78.6	7.1	14.3	100

Kappa index ranges from 0 to 1: 0 = no agreement; 0.1–0.2 = slight agreement; 0.21–0.4 = fair agreement; 0.41–0.6 = moderate agreement; 0.61–0.8 = substantial agreement; 0.81–0.99 = near-perfect agreement.

. Variables assessed include plant density per hectare, staking materials, reasons for the use of selected materials, and fertiliser use in production. Most interviewees (48.1%) proposed a planting density of 10,000 plants per hectare, with the smallest proportion (5.6%) suggesting that Kumbu is planted at 15,000 plants per hectare. The Chi-Square test of association (χ² = 0.820, p = 0.936), Cramer’s V (0.112, p = 0.936), and Cohen’s Kappa (K = 0.006, p = 0.874) show that there is no association between whether or not interviewees plant Kumbu, on the one hand, and the proposed planting density on the other.

Kumbu farmers mainly use existing trees (86.2%) for staking, while a minority (1.8%) reported that no staking is performed. The tests of association, Chi-Square (χ² = 13.289, p = 0.004), Cramer’s V (0.230, p = 0.004), and Cohen’s Kappa (K = 0.027, p = 002), show a significant association between the staking materials and whether or not interviewees plant Kumbu. Two main reasons are provided by interviewees for the choice of staking materials, namely availability (86.2%) and affordability (11.1%). A majority of the interviewees who cultivate Kumbu (94.2%) do not use any fertiliser in cultivation, and there is a significant association between fertiliser use and interviewees’ cultivation of Kumbu (χ² = 14.520, p = 0.000; V = 0.160, p = 0.000), with Cohen’s Kappa statistic (k = 0.063, p = 0.000) indicating a slight association between the independent and dependent variables. Where fertiliser use is practised in Kumbu cultivation, a majority (78.6%) of interviewees use organic fertilisers, with a further 14.3% using both organic and inorganic fertilisers. No association exists between the dependent and independent variables.

Among the few (5.8%) interviewees who apply fertilisers in the cultivation of Kumbu (Table 9 above), poultry manure, compost, and NPK fertilisers are their preferred choices. A majority (53.8%) of them use compost and 46.2% use poultry manure, with the main reason for using fertiliser being to increase yield (93.3%). While there is a significant association between the types of fertilisers used and the fertiliser users (χ² = 14, p = 0.001; V = 1, p = 0.001), no further associations were found.

Most farmers practice intercropping Kumbu with other crops. The intercrops include cocoyams (29.24%), maize (10.17%), cassava (10.17%), and common beans (4.66%). A significant proportion (45.76%) of the interviewees indicated that they perform intercropping with other crops as well but did not indicate the names of these intercrops. Of the farmers who do not practice intercropping (21.4%), the most prevalent reason provided for monocropping Kumbu is that its vines and leaves shade other plants (50%). A large proportion of this segment of interviewees (41.7%) also do not intercrop because they maintain wild stands of Kumbu.

3.7. State of Commercial Cultivation of Kumbu in the Mount Cameroon Region

A majority (94.3%) of interviewees do not cultivate Kumbu for commercial purposes. Among those who cultivate for commercial purposes (Figure 3), the most common channel of distribution is from farm to market, otherwise known as direct marketing (38.46%), followed by retail marketing where the produce is sold to retailers (26.92%). Some Kumbu is processed (cooked) before sale (19.23%).

3.8. Perceptions of Kumbu Consumption in the Mount Cameroon Region

Thematic analysis of interviewees’ perceptions of Kumbu consumption (

Table 10. Cross-tabulation of interviewees’ perceptions of eating Kumbu.

Do You Like Eating “Kumbu”		Reasons for Liking Kumbu			Total	Chi-Square	Cramer’s V	Kappa
		Good Taste	Good Tuber Colour	Others
Yes	Count	203	26	45	274	15.132, p = 0.001	V = 0.232, p = 0.001	0.054, p = 0.010
	%	74.1	9.5	16.4	100
No	Count	1	1	5	7
	%	14.3	14.3	71.4	100
Total	Count	204	27	50	281
	%	72.6	9.6	17.8	100

) shows that a total of 281 interviewees were evaluated on whether or not they liked eating Kumbu and why. A majority (74.1%) of this population segment reported that they like eating Kumbu because of its good taste. Some (9.5%) reported that they like eating Kumbu because of its good tuber colour. There was a strong association between whether or not interviewees like eating Kumbu and the reasons for this choice (χ² = 15.132, p = 0.000; V = 0.232, p = 0.001).

3.9. Kumbu Seed Preservation and Pest Incidence in the Field and in Storage

Different methods are used to preserve Kumbu seeds for the next season’s planting. Results showed that a majority of interviewees who shared their knowledge on this leave the seeds for the next season on the farm (58.1%), while others store the harvested tubers in a dry place at home (40.1%). The perceptions of interviewees regarding pest incidence during the planting and storage phases of Kumbu cultivation are presented in

Table 11. Interviewees’ perceptions on pest incidence in planting and storage of Kumbu in the study site.

Parameter	Frequency	Percent	Valid Percent	Cumulative Percent
Pests in planting
Yes	163	28	28.3	28.3
No	145	24.9	25.2	53.5
No idea	268	46	46.5	100
No response	7	1.2
Total	583	100
Pest in storage
Yes	110	18.9	18.9	18.9
No	189	32.4	32.5	51.5
No idea	282	48.4	48.5	100
No response	2	0.3
Total	583	100

. Similar to other crops, Kumbu is susceptible to pests and diseases. Notably, 46.5% of respondents reported having no knowledge of the pests affecting Kumbu during the planting stage, while 28.3% indicated awareness of various pests encountered during this phase. In terms of storage, 48.5% of interviewees were unaware of the pests that may affect Kumbu, with only 18.9% expressing familiarity with pest issues during the storage period (refer to Table 11).

4. Discussion

4.1. The State of Knowledge of Indigenous Yam Varieties Is Low and Possibly Declining

Although yam production is somewhat mainstream in West and Central Africa, the more popular species are Dioscorea alata, D. rotundata, D. cayenensis, and D. dumetorum [11]. These mainstream species dominate agricultural systems, overshadowing indigenous varieties, which are often underutilised and poorly understood. These authors [11] reported a lot of linguistic polymorphisms in the naming of yams, which was attributed to morphological differences possibly arising from polymorphism. Nonetheless, most farmers are only aware of these mainstream yam types. Indigenous yams are often unknown in the community. In the current study, about 69.3% of the interviewees know traditional yam varieties (Kumbu) of the Bakweri and Bomboko ethnic groups, and of those who expressed knowledge, just 68.5% of the interviewees have ever seen these indigenous yam varieties. This limited exposure underpins the declining prominence of indigenous yams in local agricultural practices. According to Dumont et al. [14], there are 17 wild yam varieties in Cameroon with about 10 edible ones, including D. hirtiflora Benth, D. semperflorens Uline, D. praehensilis Benth, D. mangenotiana Miège, D. burkilliana Miège (KeKe), D. minutiflora Engl, as well as three uncharacterised species. The Baka pygmies in Cameroon are reported [13] to practice paracultivation of these indigenous varieties, a practice that allows the maintenance of wild stands and the gene bank. According to Yasuoka [19], the “wild” yams exploited by the Baka of Cameroon appear to have been deliberately established and maintained. On the contrary, knowledge of indigenous yams among the Bakweri people in the southwest region remains sparse, emphasising a significant gap in comprehending the role of Kumbu in local food systems and its potential contributions to food security and biodiversity conservation. In an extensive search of online archives, only the work of Laird et al. [15] proved relevant. They [13] studied the relationship between people and their habitats in terms of resource use and found that indigenous Bakweri people in Bova Bomboko, Etome, Ekonjo, Likombe, and Upper Boando use wild “Kumbu” yams, which these authors identified as Dioscorea mummularia (Fresen.), for subsistence. The paucity of knowledge of “Kumbu” among the Bakweri is also exemplified by the fact that, in the current study, a majority of the interviewees do not cultivate Kumbu due to a lack of seeds and preference for other yams. This outcome indicates systemic barricades to the integration of indigenous yams into conventional agriculture, warranting targeted interventions such as the promotion of seed technology schemes and agricultural extension awareness campaigns to promote their cultivation.

Another dimension for assessing knowledge is the naming of the species. In the study site, a high level of ignorance of these indigenous yams is exhibited by a lack of knowledge of alternative names for Kumbu. A large majority (92.6%) did not know any indigenous or alternative name for Kumbu. The rest of the interviewees, constituting 7.4%, reported ten different names for Kumbu. In further analysis, these names increase to 14—it is not clear, for example, if Nyake, Etoh, Etale, Jaseh, or Jeme refer to the same or different yam variety or species. This variation in naming yams places accentuation on their taxonomic complexity, reflecting a broader challenge in accurately identifying and classifying these varieties. Further studies using molecular techniques are recommended to resolve these apparent differences and establish a more robust classification system. This reflects a situation that is prevalent in the naming of yam varieties in the region. This scenario mirrors Oben et al. [11], who have shown that there is high linguistic polymorphism in the naming of established yam varieties. Such polymorphism is likely even more pronounced for lesser-known indigenous yams like those of the Bakweris, given their limited cultivation and declining traditional knowledge. While molecular analyses sometimes align with classifications based on morphological characters, as Mignouna et al. [20] have shown, Siadjeu et al. [21] reported that high variation can exist within a species—in this case, D. dumetorum—sufficiently high to result in distinct populations. This implies that Kumbu may exhibit similar intra-species variation, further complicating their taxonomy and promoting the vitality of combining molecular techniques with traditional knowledge.

Another indication of the paucity of knowledge on Kumbu was found to be the characterisation of the different Kumbu varieties. Interviewees only reported differences between two Kumbu types, namely Nyake and Mwende, based mainly on tuber colour, surface smoothness, the direction of tuber growth, and taste. These distinguishing criteria align with consumer preference attributes reported by Barlagne et al. [22], suggesting that sensory and physical traits play a significant role in the perception and selection of yam varieties. Cross-tabulation results revealed that, regardless of the Kumbu type, the primary reason for preference was good taste, and this association between the preferred Kumbu type and the reason for preference was statistically significant. This finding underscores the importance of taste as a critical determinant of consumer acceptance, consistent with its role as a major screening criterion for new yam varieties [23]. Taste preferences are often shaped by cultural and culinary practices, which may explain why it emerged as the dominant factor in this study. Moreover, the preference for Nyake and Mwende could be fanned by their adaptability to local growing conditions, which may enhance their flavour profile and quality.

4.2. Kumbu Farm Sizes Are Typically Small Due to Lack of Seeds and Long Juvenile Periods with Many Stands in the Process of Domestication

Of the 38.4% of interviewees who cultivate Kumbu, farm sizes are typically small, with a large majority having less than five stands of Kumbu, mainly due to a lack of seeds and long juvenile periods. These constraints highlight the challenges associated with the adoption and expansion of indigenous yam varieties, reflecting both a lack of resources and limited agronomic knowledge about Kumbu. Considering that the average yam farm size in the Fako Division for mainstream yam varieties is 0.43 ha [24], production units of Kumbu consisting of just five stands are minute. As is the case with emergent or neglected crops, a significant proportion of the crop is harvested from volunteer stands (wild stands domesticated in farms). This practice mirrors the early stages of crop domestication, where farmers rely on naturally occurring plants as a resource base for cultivation. An antecedent to this can be found in the domestication of D. abyssinica Hochst. ex Kunth or D. praehensilis in Ethiopia, where it was reported that most farms comprised a mixture of domestic stands and stands in the process of domestication, with the domestication process taking sometimes as long as six years [25]. This suggests that the domestication of Kumbu in the Mount Cameroon Region may follow a similar trajectory, with wild stands gradually contributing to the cultivated population over time. The reliance on volunteer stands and the slow domestication process highlights the need for research into propagation techniques and agronomic practices tailored to Kumbu.

Seed availability is also a determining factor for the level of cultivation of this indigenous yam variety as most interviewees reported that a lack of seeds is a reason why they depend on farmer-saved traditional seeds, with some acquired from neighbours. This informal seed system throws light on a vital limitation in the production of Kumbu as it lacks the structured distribution and quality control mechanisms associated with improved seed systems. Azeteh et al. [13] have shown that the lack of improved seeds—alongside labour demand, pests, and diseases—is a major limitation to mainstream yam production. Similar findings have been reported by Mondo et al. [26] in the Eastern Democratic Republic of Congo. Therefore, this constraint is expected to be more severe in neglected yams like Kumbu. Farmers of Kumbu in the Mount Cameroon Region typically conserve seed tubers for the next season by preserving on the farm, either left underground or in barns, and a majority have no idea what the ideal planting size for a seed tuber should be. Storage of the seed tubers underground is a common practice in yam-producing areas and has been reported by other authors, for example, Eteiere and Bhat [27] in a study in Nigeria; additionally, an experimental study that compared shed storage to underground storage methods showed that yam tubers stored underground preserved better due to lower humidity [28]. Such findings suggest that while underground storage is effective, farmers would benefit from education on best practices for seed tuber preservation and planting techniques.

Among the interviewees, 38% had some experience in Kumbu cultivation, with most reporting over 15 years of experience. While this long-term engagement indicates a degree of traditional knowledge retention, the small proportion of experienced farmers suggests that overall knowledge of the conservation of Kumbu within the population is low. This may be attributed to the marginalisation of Kumbu in favour of mainstream yam varieties, as well as the lack of institutional support for indigenous crop cultivation. Promoting knowledge-sharing platforms, farmer training programs, and research on Kumbu’s agronomic practices could help revitalise interest and expertise in its cultivation.

4.3. Agronomic Practices in Kumbu Cultivation Are Similar to Other Yams in the Region but at a Reduced Scale

There is a strong association between the professed knowledge of the planting distance of Kumbu and the proposed distances, with a majority of those who grow Kumbu proposing a distance of 1 m × 1 m for a stand density of 10,000 m² per hectare. This spacing mirrors the recommendations for other Dioscorea species grown in the region and is likely adopted by farmers based on familiarity rather than specific knowledge of Kumbu’s agronomic requirements. This is the spacing recommended by the IITA and generally adopted by yam farmers in Cameroon, although other studies suggest that this high (1 m × 1 m) spacing may result in larger tubers but lower overall yield. Vashi et al. [29] report a revised distance of 60 cm by 45 cm for better yield in D. alata in pooled analyses over two seasons. Similarly, Suja et al. [30] found that D. rotundata performed best with a seed tuber size of 300 g and a spacing of 90 cm × 90 cm. Given that Kumbu may represent a distinct species or a species complex, further research is essential to determine the optimal spacing for maximising yield and tuber quality.

Like mainstream yam species, the cultivators of Kumbu in the Mount Cameroon Region mainly use trees retained on the farm for staking due to the fact that these are readily available and affordable. Staking helps increase the leaf biomass and photosynthesis, reduces soil-borne diseases and, hence, increases production. However, as Danquah et al. [31] demonstrated, the benefits of staking can be species-dependent, with some yam species performing well even with minimal or no staking. This suggests that further studies are needed to evaluate whether Kumbu requires staking or if alternative methods, such as trellising or reduced staking, could be more effective. In the current study, fertiliser use in Kumbu production is still minimal, and where it is practised (5.8%), organic manure is preferred. This limited use of fertilisers appears to be incidental, driven by availability rather than a deliberate farm fertility management strategy. It is unclear how this affects yields in the region; however, minimal or zero fertilisation farming is common among the indigenes of the Mount Cameroon Region where the soils are of volcanic origin and naturally fertile. Despite this, some degree of fertilisation is recommended for improved production of yams, for example, Agbede et al. [32] recommended synergistic use of oil palm bunch ash and poultry manure at 10 tons/ha; Diby et al. [33] stressed the importance of soil organic matter in yam production systems, a lack of which the authors reported, which cannot be remedied by inorganic fertilisers. Danquah et al. [34] reviewed several fertility options for yam production and proposed sustainable fertility options that include integrating legumes into yam farming systems for sustainable production. These findings suggest that research into sustainable fertilisation practices for Kumbu, tailored to the region’s unique soil conditions, could significantly improve yield and support long-term productivity.

Intercropping Kumbu with cocoyam, maize, cassava, and common beans is practised, probably because its cultivation is still rudimentary and there is a need for these other crops that are more mainstream for sustenance. Intercropping provides food diversity and, in the case of common beans, offers the additional benefit of nitrogen fixation, which contributes to integrated soil fertility management [34]. For those interviewees who do not practice intercropping (21.4%), the main reason provided includes the shading effect of Kumbu foliage on the understory crops. This suggests that for these farmers, the co-crops are considered more important than Kumbu, further reinforcing its status as a subsistence crop. Indeed, 94.3% of the interviewees cultivate Kumbu mainly for subsistence, further indicating that it is not a mainstream crop.

Another consideration in Kumbu production is pest and disease management. Few of the interviewees (28.3%) are aware of the pests that are prevalent at the production stage, and even fewer know that pests of Kumbu in storage exist. Once again, this lack of awareness reflects a broader knowledge gap regarding the management of this indigenous crop. Like all other yams, Kumbu is susceptible to pests and diseases. Tariq et al. [35] reviewed the diseases of yams and their management and reported major diseases like anthracnose, leaf spot, yam wilt, dry rot, and crazy root syndrome. Given that the Fako Division is a major yam-producing zone, it is likely that Kumbu is also vulnerable to these diseases, which could further limit its productivity.

5. Conclusions

The results paint a picture of Kumbu as either a wild yam in the process of domestication or a wild yam in the process of extinction, for which knowledge of its virtues and agronomic practices in the study site—as well as in the published literature—is sparse. This study unveils key findings relating to the cultivation of Kumbu. It revealed that 38.4% of interviewees grow Kumbu, with 69.3% citing seed unavailability as a major barrier. Handling seed production and provision through propagation research could enhance cultivation and diversify crops. It is recommended that research on agronomic practices like mini-sett techniques, fertilisation options, and planting distances of Kumbu should be conducted as it will clarify the better conditions for Kumbu cultivation. This research draws from best practices that are applied to mainstream yam species. Secondly, 68.1% of interviewees could not distinguish between Kumbu types. Although several names are known for Kumbu that represent different “types”, this study presents evidence that this could be a case of genetic and linguistic polymorphism, requiring further molecular studies for proper identification and characterisation. As a perspective, molecular characterisation studies should be conducted on Kumbu “types” as it would resolve the classification and naming issues of Kumbu and preserve its diversity. About 30.7% of non-cultivators prefer other yam types, accentuating the importance of promoting Kumbu’s nutritional and cultural value through education and extension services. This study indicates that there is a decline in indigenous knowledge and limited agronomic practices specific to Kumbu, suggesting the necessity for tailored research in cultivation techniques and pest management. Low levels of adoption of Kumbu into farming systems suggest the need for valuation and research on better farming methods ranging from seed production to fertilisation. This research contributes to indigenous knowledge conservation and has the potential to improve the conservation and adoption of Kumbu into mainstream agricultural policies. This has the potential, therefore, to improve crop diversification and food security and protect biodiversity in the region. Further studies on the adaptability of Kumbu to climate change, especially rainfall and temperature variability, are recommended. In all, our results are significant in that they provide baseline data on the state of indigenous knowledge of Kumbu cultivation, which could be utilised both for further research and to influence agricultural policy.