Phenotypic Variability and Evaluation of Yam (Dioscorea spp.) Landraces from Southwest Ethiopia by Multivariate Analysis

 

Tewodros Mulualem^1,2, Firew Mekbib², Shimelis Hussein³, Endale Gebre⁴

¹Jimma Agricultural Research Center, P.O. Box 192, Jimma, Ethiopia

²Haramaya University, School of Plant Sciences, P.O. Box 138, Dire Dawa, Ethiopia

³African Centre for Crop Improvement, School of Agriculture, Earth and Environmental Sciences,

University of Kwa Zulu-Natal, Private Bag X01, Scottsville, 3209, Pietermaritzburg, South Africa.

⁴ Ethiopian Institute of Agricultural Research, P.O. Box 2003, Addis Ababa, Ethiopia

 

ABSTRACT:

Yams (Dioscorea spp.) are food security and socioeconomic importance crop in different areas of the world. Although it is cultivated and consumed in sub- Saharan Africa, still neglected by scientific research and development program in many countries including in Ethiopia. To fill in the knowledge gaps, the present study conducted at Jimma Agricultural Research Center during 2015. The objectives of the study were to characterize and assess the level of diversity within farmers and reference collection of yam landraces collected from major growing areas of Southwest Ethiopia. Thirty-six landraces of yam tested by using 6 x 6 simple lattice design with two replications. Data on 32 qualitative morphological traits collected and subjected to multivariate analyses. Cluster analysis based on qualitative characters revealed seven distinct clusters with varying sizes and presence of variability, based on their foliar and subterranean traits which will be highly useful in the genetic improvement. The result of the Shannon-Weaver diversity index (H`) indicated, the existence of a high level of diversity among Dioscorea spp. landraces based on the frequency distribution of phenotypic traits that considered. The results of PCA indicated that characters that have a good contribution to the variability. The first seven principal components explained 88.4% of the total variation, while PC-I and PC-II accounted 55.30% of the total variability. Thus, this utmost phenotypic variability between landraces is vital for hybridization and to produce potential and meaningful hybrids and desirable segregants of yams in Ethiopia.

 

 

 

 

INTRODUCTION:

Yam is a multi-species crop that belongs to the genus Dioscorea family (Coursey, 1967, Tamiru et al. 2007). The genus Dioscorea belongs to the monocotyledonous family Discoreaceae, the most imperative family within the order Dioscoreales (Burkill, 1960, Ayensu and Coursey, 1972, Dansi et al. 1999). It is grown in Africa, India, Southeast Asia, Australia and tropical America with more than 600 species (Jayasurya, 1984, Wilkin, 1998, Mignouna et al. 2002, Mulualem and Mohammed, 2013). All species are the tropical origin, are grown for their edible tubers and bulbils. About ten species are food yams and cultivated as staples for millions of people throughout the tropics (Hahn and Hozio, 1993, Dansi et al. 2013, Sesay et al. 2013).

 

Among from all species, Discorea alata, Discorea bulbifera, Discorea cayenensis and Discorea rotundata complex are the most widely cultivated and have real economic significance in Africa (Lebot, 2009, Asiedu and Alieu, 2010, Norman et al. 2012). Even though yams are cultivated mainly in most tropical countries, the most important species are originated from Ethiopia (Coursey, 1967, Zeven and De Wet, 1982; Tamiru et al. 2011), and grown in African (Terauchi et al. 1992). Miege and Demissew (1997) described eleven Dioscorea species grown in Ethiopia. Besides, 23 indigenous yam types belonging to four Dioscorea species such as, D. bulbifera (aerial yam), D. alata (water yam), D. cayenensis and D. rotundata Complex are widely distributed in major growing areas of the country for food, medicinal use and to fill economic gaps during the absence of other crops in the field (Vavilov, 1951, Coursey 1967, Hildebrand, 2002). Furthermore, various Dioscorea species growing in the complex farming system as cultivated and wild forms with cereals and other root and tuber crops (Edwards, 1991).

 

Despite its contribution to food security, as medicinal, social and economic value, yams in Ethiopia have been poorly investigated and farmers’ distinguished their cultivated landraces on the bases of morphological traits by using their indigenous knowledge. Conversely, the existence of different vernacular names for the same cultivar of the species has created problems to classify landraces while avoiding duplicates for conservation (Karamura, 1998; Mulualem, 2008).

 

Morphological characterization is the first step in grouping and description of the crop (Smith and Smith, 1989). Various characterization techniques have been successfully used to classify and measure the pattern of phenotypic diversity in the relationship of crop landrace collections of economically important traits. Genetic diversity can be described by four levels of organization that is, among species, among populations, within populations and within individuals (Hunter, 1996). The use of multivariate statistical methods is an imperative strategy for classifying landrace, ordering variability for a large number of landraces or analyzing genetic relationships among breeding materials (Mohammadi and Prasanna, 2003). To enhance productivity and to promote the uses of yam landraces, better understanding on the patterns of variability and grouping of available landraces are needed to boost yam production, which minimizes the poverty and improves the livelihood of rural households. To utilize the indigenous yam genetic resource in Ethiopia and to analyze the diversity present in on farm, characterization, evaluation of the existed landraces and using multivariate analyses is essential. Cognizant to these facts, the present study was designed to constitute landrace groups based on morphological resemblance, assess the level of diversity within yam collection, identify the key traits contributing to the variability, and to find superior landraces for the future breeding program.

 

MATERIALS AND METHODS:

Study area:

The study conducted at Jimma Agricultural Research Center (JARC). The center located at latitude 7^o 40.00' N and longitude 36^o 47’.00’ E with an altitude of 1753 m.a.s.l. The area received mean annual rainfall of 1432 mm with the maximum and the minimum temperature of 26.5⁰C and 12.00 ⁰C, respectively. The soil of the study area is Eutric Nitosol (reddish brown) with p^H of 5.3.

 

Experimental materials, design, and management:

The landrace studied consisted of 36 yam landraces collected from seven districts of Jimma, Sheka and, Bench-maji zones of Southwest Ethiopia. The experiment was laid out in 6x6 simple lattice design with two replications. Single row plots, with each row 7m long and spacing of 1.5m between rows, and 1m between plants within a row used. The layout and randomization of the field plots carried out based on the standard procedures outlined by Cochran and Cox (1957). Tubers of the same size and just started sprouting were used as planting material. Planting done on the ridges in the row. One month after planting, after the crop well established, the seedlings earthed up, frequent weeding and other agronomical operations were carried out consistently to entire treatments according to the farmer’s practices. Individual plants supported by the individual stake of dried coffee about 3.5-4.5 meters long above ground to encourage good canopy development. Five plants of the row from the middle were used for data collection and harvest.

 

Table 1. Descriptions the 36 yam landraces used for the study.

S. No.	Name of landrace	Zone	District	Latitude	Longitude	Altitude
1	59/02	Jimma	Mana	07040’37N	036049’10E	1718
2	68/01	Jimma	Dedo	07030’63N	036053’45E	1784
3	6/02	Bench maji	Sheko	06059’66N	035034’11E	1728
4	75/02	Jimma	Kersa	07040’43N	036048’76E	1734
5	3/87	Jimma	Manna	07040’58N	036048’75E	1731
6	56/76	Jimma	Manna	07041’89N	036048’06E	1837
7	54/02	Bench maji	Sheko	07002’03N	035032’77E	1892
8	46/83	Jimma	Dedo	07031’28N	036053’59E	1771
9	08/02	Jimma	Kersa	07040’46N	036048’79E	1740
10	116	Jimma	Dedo	07031’28N	036053’63E	1683
11	01/75	Sheka	Yeki	07011’30N	035026’22E	1171
12	06/83	Jimma	Dedo	07031’32N	036053’64E	1692
13	17/02	Sheka	Yeki	07011’27N	035026’26E	1176
14	07/03	Jimma	Dedo	07031’50N	036053’60E	1733
15	45/03	Jimma	Mana	07041’86N	036048’08E	1810
16	27/02	Jimma	Sekachekorsa	07035’06N	036041’91E	1877
17	37/87	Jimma	Mana	07041’87N	036048’13E	1940
18	10/002	Bench maji	Sheko	07002’91N	035029’76E	1668
19	76/02	Jimma	Kersa	07040’64N	036048’84E	1728
20	06/2000	Jimma	Sekachekorsa	07035’43N	036041’86E	1850
21	7/83	Jimma	Sekachekorsa	07035’06N	036041’91E	1898
22	58/02	Sheka	Yeki	07011’22N	035026’25E	1192
23	39/87	Jimma	Sekachekorsa	07035’42N	036042’94E	1885
24	32/83	Jimma	Shebesombo	07026’74N	036024’01E	1372
25	24/02	Jimma	Shebesombo	07026’75N	036024’07E	1379
26	2/87	Jimma	Shebesombo	07026’76N	036024’12E	1365
27	60/87	Sheka	Yeki	07011’72N	035026’48E	1199
28	15/2000	Bench maji	Sheko	07004’13N	035037’74E	1320
29	34/87	Jimma	Dedo	07031’37N	036053’44E	1911
30	21/02	Jimma	Sekachekorsa	07036’48N	036045’09E	1785
31	57/76	Bench maji	Sheko	07002’88N	035029’74E	1654
32	0001/07	Jimma	Shebesombo	07026’74N	036024’12E	1367
33	0004/07	Jimma	Kersa	07040’55N	036048’75E	1741
34	7/84	Bench maji	Sheko	07002’88N	035029’74E	1661
35	7/85	Sheka	Yeki	07014’30N	035026’17E	1173
36	06/2001	Bench maji	Sheko	06059’69N	035034’09E	1387

 

Morphological data collection:

A descriptor of yam (Dioscorea spp.) developed by Bioversity International (IPGRI, 1999) followed for data collection. A total of 32 qualitative characters measured for this study. Data recorded from the middle of five plants of the row, and the average value used for statistical analysis. Munsell color chart used for color identification. The lists of qualitative characters presented as follow:

Leaf color: 1= yellow-green, 2= pale-green,

3=dark-green, 4= purplish-green, 5=purple,99=other

Leaf vein color upper surface: 1= green,

2= yellow-green, 3=pale-purple, 4= purple, 99=other

Leaf vein color lower surface: 1= green,

2= yellow- green, 3= pale- purple, 4= purple, 99= other

Leaf margin: 1=Entire 2= Serrate

Leaf margin color: 1= green, 2= purple,

3= yellow-green, 99= other

Leaf shape: 1= ovate, 2= chordate 3= chordate-long, 4=chordate-broad, 5= sagittate

Leaf apex shape: 1= obtuse 2= acute, 3=emarginated

Hairiness of leaf surface: 0= present, 1= absent

Leaf position: 1= alternate, 2= opposite

Leaf size: 1= Small, 2= Medium, 3= Large

Leaf density: 1= Low, 2= Medium, 3= High

Leaf lobations: 1= shallow, 2=deep

Waxiness: 0= absent 1= present

Tip color: 1= light-green, 2= dark-green, 3= purple,

4= red, 99=other

Petiole color: 1= all green with purple base,

2= all green with purple leaf junction, 3= all green with purple at both ends, 4= all purplish green with purple base, 5= all purplish green with purple leaf junction, 6= all purplish green with purple at both ends,7= green, 8= purple, 9= brownish green, 10= brown, 11= dark brown, 99=other.

Petiole wing color: 1= green,

2= green with purple edge, 3= purple, 99= other

Twinge direction: 1= clockwise, 2= anticlockwise

Vine color: 1= yellowish, 2= green,

3= light-green, 4=purple, 99=other

Wings on vine: 1=present, 2= absent

Wing color: 1= green, 2= green with purple edge,

3= purple 99= other

Spines: 1= present, 2= absent

Spine shape: 1= straight, 2= curved

Color at spine base: 0= present, 1= absent

Spine on the vine base: 0= none, 3= Few, 7=many

Tuber shape: 1= round, 2= oval, 3= oval-oblong,

4= cylindrical, 5= flattened, 6= irregular, 99= other

Tuber skin color: 1= grayish, 2= light-brown,

3=dark-brown, 4= 0ther

Tuber branching: 0= none, 1= slightly branched,

2= branched, 3= highly branched

Tuber surface texture: 1= smooth, 2= rough

Tuber flesh color: 1= white, 2= yellowish-white,

3= yellow, 4= orange, 5= light-purple, 6= purple

7= purple with white, 8= white with purple, 9= outer purple/ inner white, 99= other

Hairiness of tuber surface: 0= absent, 1=small, 2=medium, 3= large

Flower: 0= present, 1 absent

Flesh color at central transverse cross section: 1= white, 2= yellow- white, 3= yellow, 4= orange, 5= light- purple, 6= purple, 7= purple with white, 8= white with purple, 9= outer purple/inner yellowish, 99 = other

 

Statistical analysis:

Cluster analysis based on qualitative traits:

Clustering was done using the Statistical Analysis System (SAS) package (version 9.0 of SAS Institute Inc, 2000) based on Un-weighted Pair Group Methods with Arithmetic average (UPGMA). The number of clusters was determined by looking into two statistics namely, pseudo F and pseudo T². The number of the cluster decided where the pseudo F statistics combined with the small value of pseudo T² and large pseudo T² statistical for the next cluster fusion (Milligan and Cooper, 1985). The dendrogram constructed by UPGMA (Afifi and Clark, 1990).

 

Frequency distribution and Shannon-Weaver diversity index (H’):

The frequency distribution is a systematic way of ordering a set of data from the lowest to the highest value showing the number of occurrences (frequency) at each value or range of values. In the latter, the distribution is called a relative frequency distribution. The Shannon-Weaver diversity index was used to determine the diversity of the landraces collected from different districts of Southwest Ethiopia by using the frequency of distributions and the number of phenotypic classes (Hennink and Zeven, 1991). The index defined as:

 

 

Where p_i is the proportion of the total number of individuals (landrace) in the i^th class and, S is the number of phenotypic classes.

 

The Shannon Weaver index values (H’) can ranged from 0 to ~ 4.6 using the natural log (versus log₁₀). A value near 0 indicated that every species in the sample are the same. Conversely, a value near 4.6 showed the numbers of individuals evenly distributed between the species (Hennink and Zeven, 1991).

 

Principal component analysis:

The principal component analysis performed by using a correlation matrix by employing SAS version 9.0 (SAS, 2000). The objective of this analysis was to reduce the observed variables into a small number of principal components that accounted for most of the variance in the observed variables. The first step in PCA is calculating the eigen values, which define the amount of total variation that displayed on the PC axes. The first PC groups most of the variability present in the original data relative to all remaining PCs. The second PC explains most of the variability not summarized by the first PC and uncorrelated with the first, and so on (Mohammadi and Prasanna, 2003).

 

RESULTS AND DISCUSSION:

Cluster analysis:

Grouping of landraces based on their similarity is essential. In the present study, this approach was adopted to cluster the landraces into seven different groups with different sizes based on 32 qualitative characters (Table 2). A dendrogram summarize genetic similarly among 36 Dioscorea landraces based on qualitative characters are given in Figure 1. The number of landraces belonging to each cluster was diverse and varied from one in cluster VII to 11 in cluster I and II (Table 2).

 

Cluster I was the largest and consisted 11(30.55%) of the total landraces. Of these, nine from Jimma zone (3 each from Dedo and Seka chekorsa, two from Kersa and one from Shebe- sombo), and one landrace each from Bench-Maji and Sheka collections. Landraces grouped under this cluster have mainly identified by dark-green leaf, green vein on upper and lower surface, entire and yellow-green leaf margin, chordate leaf, acute leaf apex, opposite leaf position, medium leaf size, medium leaf density, light-green tip and petiole, cylindrical tuber shape, light-brown and slightly branched tuber, large hair rough tuber surface with white with purple flesh color at the central transverse.

 

Likewise, Cluster II also accommodated 11 landraces (30.55%), six from Bench-Maji and five from Jimma collections. Landraces failed into this cluster different from cluster I by having yellow-green vein on upper surface, obtuse leaf apex, small leaf size, all green with purple edge petiole, green vine, irregular and branched tuber with medium hair, rough tuber surface and light -purple flesh color at the central transverse. Besides, four landraces (11.11%) grouped under cluster III, all from Sheka and Jimma collections. They typically possessed yellowish and dark- green leaf, yellowish and green leaf vein on upper surface, yellow-green leaf margin and leaf vein on lower surface, all green with purple at both ends of petiole, obtuse leaf apex, purple tuber flesh, dark- brown tuber skin and purple with white flesh color at the central transverse.

 

Three landraces (8.33%) grouped under cluster IV, all from Jimma (2 from manna and one from Shebe-sombo) zone. Predominantly, landraces differ from the other clusters by having shallow and deep leaf lobs, dark-green tip color, all green with purple at both ends petiole color, clockwise and anti-clockwise twining direction, light-green vine, green wing and flattened tuber shape. Some landraces categorized under this cluster, produce flower at maturity. Finally, seven landraces were (19.44%) grouped under cluster V, VI and, VII, two from Sheka and five from Jimma collection. All landraces typically possess the combination of all characters described under the above four clusters. The results indicated landraces collected from diverse agro-ecological areas of Southwest Ethiopia; nevertheless, landraces from the same or different zones fall into different clusters and showing different genetic make-up. Thus, it may be of considerable importance to enlarge the genetic bases of yams by a sustainable and continual collection of landraces throughout the major growing areas of the country for further genetic enhancement of the crop. Dewey and Lu, (1959) and Kifle (2006) who further confirmed that while selecting landraces from a particular cluster, the inter-cluster distance and cluster mean performance of traits should be taken into consideration.

The cluster means of 19 quantitative characters were assessed based on clusters formed by 32 qualitative characters and showed that cluster VII had the highest value in most of the characters that considered in this study; for example, longest leaves, longest leaf lobs, longest vine, longest tip and widest leaves, highest number of tuber and vine hill^-1and vine fresh weight (Table 3). Cluster IV showed high performance among the characters that considered. For example had highest petiole length, distance between lobs, tuber length, number of internodes vine^-1, number of vine hill^-1, tuber fresh weight, tuber dry weight, and harvest index. Cluster I showed the highest vine fresh and dry weight. Cluster III had the highest days of maturity. However the rest of cluster had the average performance of the traits considered.

 

 

Table 2. Clusters of Dioscorea spp. landraces based on qualitative traits

Clus-ters	Number of land races in each cluster	Serial number	Name of landraces in each cluster	Major characteristics
I	11	29,36,19,26,16,4, 8,30, 12, 27 and 21.	34/87, 06/2001, 76/02, 2/87, 27/02, 75/02, 46/83, 21/02, 06/83, 60/87, and 7/83	LC=dark green, LVCUS= green, LVCLS= green, LMC=yellow green, LM= entire, LS= chordate, LAS= acute, LPO= opposite, LSi=medium, LD=medium, LLO= shallow, Tic= light green, PC=light green with purplish leaf junction, PWC=green with purple edge, Twdir=clockwise, VC= yellow green, Wic= green, Tsh= cylindrical, Tsc= light brown, TFC=light brown, TBr.= slightly branched, TStex= rough, HoTs= large and FCCS= white with purple.
II	11	23,24,3, 34, 28, 31,7,18,25,2 and 5.	39/87, 32/83, 6/02,7/84, 15/2000, 57/76, 54/02, 10/002, 24/02, 68/01and 3/87.	LC=dark green, LVCUS= yellow green, LVCLS= green, LMC=yellow green, LM= entire, LS= chordate, LAS= obtuse, LPO= opposite, LSi =small, LD=medium, LLO= shallow, Tic= light green, PC= all green with purple edge, PWC= green, Twdir= clockwise, VC= green, Wic= green, Tsh= irregular, TFC=light brown, Tsc= light brown, TBr.= branched, TStex= rough, HoTs= medium and FCCS= light purple.
III	4	22, 35, 10 and 20.	58/02, 7/85, 116 and 06/2000. .	LC=yellowish and dark green, LVCUS=yellow and green, LVCLS= yellow green, LMC=yellow green, LM= entire, LS= chordate, LAS= obtuse, LPO= opposite, LSi=small, LD=medium, LLO= shallow, Tic= light green, PC= all green with purple at both ends, PWC=green, Twdir= clockwise, VC= light green, Wic= green, Tsh= oval oblong, TFC=purple, Tsc= dark brown, TBr.= branched, TStex= rough, HoTs= medium and FCCS= purple with white.
IV	3	6, 32 and 1.	56/76, 0001/07 and 59/02	LC=pale green, LVCUS=yellow green, LVCLS= yellow green, LMC= green, LM= entire, LS= cordate, LAS= obtuse, LPO= opposite, LSi=medium, LD=medium, LLO= shallow and deep, Tic= dark green, PC= all green with purple at both ends, PWC=green, Twdir= clockwise and anti clockwise, VC= light green Wic= green, Tsh= flattened, TFC=outer purple inner white, Tsc= dark brown, TBr.= branched, TStex= rough, HoTs= medium and FCCS= purple with white.
V	4	13, 14, 15 and 17.	17/02, 07/03, 45/03 and 37/87.	LC= green, LVCUS=yellow and green, LVCLS= yellow green, LMC= green, LM= entire, LS= chordate, LAS= acute, LPO= opposite, LSi=small, LD=high, LLO= shallow, Tic= green, PC= all green with purple at both ends, PWC=green with purple edge, Twdir= clockwise, VC= light green, Wic= green, Tsh= irregular, TFC=purple with white, Tsc= dark brown, TBr.= branched, TStex= rough, HoTs= small and FCCS= purple with white.
VI	2	9 and 11	08/02 and 01/75.	LC=pale green, LVCUS=yellow green, LVCLS= yellow green, LMC=green, LM= entire, LS= chordate, LAS= obtuse, LPO= opposite, LSi=medium, LD=medium, LLO= shallow and deep, PC= all green with purple at both ends, PWC=green, Twdir= clockwise and anti clockwise, VC= light green, Wic= green, Tsh= round and oval, TFC=light purple, Tsc= dark brown, TBr.= slightly branched, TStex= rough, HoTs= medium and FCCS= purple with white.
VII	1	33	0004/07.	LC= yellow green, LVCUS=yellow green, LVCLS = yellow green, LMC= green, LM= entire, LS= chordate, LAS= obtuse, LPO= opposite, LSi= medium, LD= medium, LLO= shallow, Tic= green, PC= all green with purple at both ends, PWC=green with purple edge, Twdir= clockwise, VC= green, Wic= green, Tsh= round, TFC=purple with white, Tsc= dark brown, TBr.= branched, TStex= rough, HoTs= large, and FCCS= purple with white.

LC=Leaf color, LVCUS= leaf vein color upper surface, LVCLS= leaf vein color lower surface, LMC= Leaf margin color, LM= Leaf margin, LS=Leaf shape, LAS = Leaf apex shape, LPO=Leaf position, LSi=Leaf size, LD=Leaf density, , LLO=Leaf lobation, Tic= Tip color, PC=Petiole color, PWC= petiole wing color, Twdir= twing direction, VC=vine color ,Wic= wing color, TSh=tuber shape, Tsc=Tuber skin color, TBr= tuber branching, TStex= tuber texture, TFC=Tuber flesh color, HOTs= hair on tuber surface, FCCS= flesh color at central transverse

 

Table 3. Cluster means of 19 quantitative traits of 36 Dioscorea spp. based on qualitative characters

Cluster	Characters
	LL	LW	PL	LLo	DBL	VL	IL	TiL	NIPV	NTPH
I	10.3	4.24	10.2	2.22	3.51	251.8	9.61	2.59	25.53	4.41
II	10.5	4.23	10.1	1.90	3.36	255.3	9.62	2.50	26.94	4.32
III	10.2	4.36	11.5	1.45	3.40	246.1	9.56	2.52	23.88	3.93
IV	11.6	4.60	9.93	2.78	3.66	256.4	9.75	2.87	25.81	4.13
V	9.87	3.94	9.28	1.54	3.24	265.4	9.80	2.32	28.88	4.58
VI	10.9	4.08	10.3	2.60	3.55	255.1	10.1	2.63	22.90	3.89
VII	12.6	4.80	11.9	1.68	3.15	257.0	10.7	2.29	24.55	3.80

Continued Table 3.

Cluster	Characters
	DM	NVPH	TL	TDi	VFW	VDW	TFW	TDW	HI
I	137.4	4.40	38.4	14.48	12.64	3.37	29.54	20.74	67.41
II	133.1	4.40	38.5	14.17	13.35	3.03	26.34	21.26	63.15
III	135.4	4.20	38.2	16.08	11.42	3.17	23.80	19.09	65.45
IV	149.2	4.40	38.9	15.68	9.56	3.04	39.20	22.13	79.21
V	142.5	4.20	39.2	15.62	12.89	2.91	35.30	20.40	72.51
VI	145.5	4.30	40.2	15.69	12.28	2.81	37.60	21.35	75.48
VII	136.9	4.20	42.7	15.68	15.44	3.21	31.60	18.31	66.91

LL=Leaf length(cm); LW= Leaf width(cm); PL= Petiole length (cm); LLo = Length of leaf lobe (cm); DBL= Distance between lobs (cm); VL= Vine length(cm);IL= Internodes length (cm);TiL= Tip length (cm);NIPV= Number of internodes/vine; NTPP= Number of tubers/hill; DM= Days to maturity;NVPH= Number of vine per hill;TL=Tuber length(cm); TDi= Tuber diameter (cm); VFW= Vine fresh weight (t/ha); VDW=Vine dry weight (t/ha);TFW=Tuber fresh weight(t/ha); TDW=Tuber dry weight(t/ha)and HI= Harvest index (%).

 

 

Figure 1. Dendrogram showing hierarchical clustering patterns of 36 Dioscorea spp. landraces (UPGMA) based on 32 qualitative characters

 

Comparison of yam landraces between zones and districts:

To know whether landraces restricted to the zones/districts or not, using comparison analysis was carried out between landraces and area of collection from southwest Ethiopia based on cluster analysis (Table 2). The results revealed that four landraces namely: 56/76, 59/02, 0001/07 and 0004/07 restricted to Jimma zone (Manna, Kersa and Shebe-sombo). Besides, landrace, 08/02 and 01/75 restricted to Jimma (Kersa) and Sheka zone (Table 4). The four landraces grouped under cluster IV and VII which are found mostly in the Jimma zone are absent in Bench-Maji and Sheka zones. This is due to the geographical differences between the districts. This is true for similar and dissimilar districts. Similar districts belong to the same boundary and have high possibilities to share genetic materials between farmers; however, districts not bounded by the same boundary had different landraces. However, the association hardly follow similar tendency, as the most similar zones of Jimma, Bench-Maji and Sheka, Dedo and Sheko and districts Kersa manna and Shebe-sombo were also among those located far apart. About 11(30.55%) of the total landraces grouped under cluster I found in all zones and districts except manna. This might be due to all districts have a similar agro ecology that was conducive for different landraces. This is also further indicated that landraces that found in one district/zone can adapt in another district when the possibility of exchange materials between farmers for conservation strategy to design.

 

Table 4. The distribution of yam landraces and areas of collection based on cluster analysis

Zone	Districts	Clusters							Total
		I	II	III	IV	V	VI	VII
Jimma	Dedo	3.0	1.0	1.0	0.0	1.0	0.0	0.0	6.0
	Kersa	2.0	0.0	0.0	0.0	0.0	1.0	1.0	4.0
	Manna	0.0	10	0.0	2.0	2.0	0.0	0.0	5.0
	Seka-chekorsa	3.0	1.0	1.0	0.0	0.0	0.0	0.0	5.0
	Shebe-sombo	1.0	2.0	0.0	1.0	0.0	0.0	0.0	40
Bench Maji	Sheko	1.0	6.0	0.0	0.0	0.0	0.0	0.0	7.0
Sheka	Yeki	1.0	0.0	2.0	0.0	1.0	1.0	0.0	5.0
Total		11.0	11.0	4.0	3.0	4.0	2.0	1.0	36.0

 

Phenotypic traits distribution:

Individual characters differed in their patterns of distribution and amount of variation among the 36 landraces. Most of the landraces in this study exhibited variation in foliar and subtraninnian plant parts are given in Figure 2.

 

Leaf color varied between landraces collected from the same and different districts. The dominant colors are being dark-green (61.11%), pale-green (25.00%) and, yellow-green (13.89%). No landraces with purple leaves on both surfaces observed. This may probably an adaptation for increasing the photosynthesis rate. A similar distribution found in the color of leaf vein upper surface. However, (88.89 %) of the landraces had yellow-green and (11.11%) had green leaf vein upper surface. Yellow-green and green leaf vein lower surface were observed for 83.33 and 16.67% of the landraces, respectively, with simple, entire and long petioles. Fifty percent of the landraces had green leaf margins, whereas, 27.78% had yellow-green leaf margins with an opposite leaf arrangement, while, 66.67 and 33.33% of the landraces had acute and obtuse leaf apex shape (Table 5). Most landraces (55.55%) having medium leaf size with chordate shape (66.67%), high leaf density (77.78%), shallow (86.11%) leaf lobe, light-green (63.89%) tip color. Thirty-three percent of the landraces had a purplish-green petiole with purple base, whereas, 63.89% had green petiole with a purple edge and, petiole wing color (Table 5).

 

The predominant vine color was light green (58.33%), while the remaining (41.67%) of the landraces had green vines with a few purple spots. Most of the landraces (94.44%) having clockwise twining direction at emergence. Few landraces (25%) produce spine on their vine with variable shape; 22.22% had curved, and, 8.33% had a straight shape, and highly associated with the wildness of the landraces. In wild type landraces spines distributed on the surface of vine and tuber in different amount and variable in size. Landraces collected from the same and different districts showed differences in the above vegetative plant parts (Figure 2). Besides, the variation in foliar before and after maturity observed among landraces. This also indicated the wide range of variation in different traits of yams in southwest Ethiopia. Thus, it needs strong attention in respect to conservation of yam genetic resources in Ethiopia.

 

Tuber shape of the landraces varied from irregular (36.11%) to oval (8.33%). The predominant tuber flesh color was white with purple (25.00%), followed by outer purple/inner white (22.22%), purple (19.44%) and, purple with white (13.89%) with dominant light and dark brown tuber skin color. Although, most of the landraces (44.44%), considered in this study exhibited branching tuber with rough (77.78%) and smooth (22.22%) surface. The flower of landraces predominantly produced spike type of inflorescent (25.0%) the other landraces hardly produced flower during the entire growing period. The predominant tuber flesh color at central transverse cross-section was white (38.89%). The other flesh colors observed included white with purple (19.44%) and (13.89%) of landraces produced similar color, for example, light purple, purple and purple with white flesh color. This result in line with Tamiru et al. (2006), who reported that there is a wide range of variability of tubers among Dioscorea species in south Ethiopia. Furthermore, a similar result was reported by Nebeyu (2003) in cassava (Manihot esculenta Cranz) and Dagne (2007) in taro (Colocasia esculenta).

 

 

A. The vegetative plant part

        

 

 

 

 

 

 

 

B. The reproductive plant part

             

                                         A seed of yam                                                   A seed                               Flower

 

C Tuber collected from different landraces

 

Figure 2. Vegetative, reproductive and storage organ (tubers) of yams of different landrace collections from Southwest Ethiopia.

 

The Shannon-Weaver Diversity Index (H')

Assessment of genetic diversity is an important aspect of any crop improvement program to identify high yielding landraces (Rhman and Munsur, 2009). In the present study, the Shannon-Weaver diversity index (H’) was adopted to compute the diversity of Dioscorea spp. based on the frequency distributions of 32 qualitative morphological characters. The result of ‘H’ value for all observed phenotypic characters showed a high level of diversity among 36 Dioscorea spp landraces, which ranged from 0.21 for vine twining direction to 1.64 for petiole color (Table 5). Besides, the overall mean of ‘H’ value of 0.72 confirmed the existence of phenotypic diversity among yam landraces from southwest Ethiopia. This result was also in agreement with the works of Tamiru et al. (2011) who found an average level of diversity in yam collection from South Ethiopia and Silvia et al. (2006) in yam collection from Colombian.

 

 

 

High ‘H’ value indicates a relatively high level of diversity and evenly distribution of landraces (Hennink and Zeven, 1991; Kifle, 2006). Furthermore, a lower level of diversity was noticed on foliar qualitative traits (Table 5). This showed that subtraninian qualitative traits had a greater influence on the phenotypic diversity among the Dioscorea spp. landraces than the foliar traits. On the other hand, mono-morphism observed in some foliar characters; leaf margin, leaf position, wing on vine revealing that the contribution of these characters to the diversity was low and omitted these characters for principal component analysis. The low level of diversity may also indicate the narrow genetic base of the plant and the low level of sexual reproduction in yam. These results indicated the need for further study based on genetics (molecular) investigation on yam.

 

 

 

 

 

 

Table 5. Frequency distribution and the Shannon Weaver diversity indices (‘H’) of 32 qualitative traits of Dioscorea grown at Jimma, 2015.           

S. No.	Qualitative character	Index and description adopted	Frequency (%)	H’
1	Leaf color	Yellow green	13.89	0.92
		Pale green	25.00
		Dark green	61.11
2	Leaf vein upper color	Yellow green	88.89	0.35
		Green	11.11
3	Leaf vein lower color	Yellow green	83.33	0.45
		Green	16.67
4	Leaf margin color	Green	50.00	1.03
		Purple	22.22
		Yellow green	27.78
5	Leaf margin	Entire	100.00	0.00
6	Leaf shape	Ovate	13.89	0.86
		Chordate	66.67
		Sagittate	19.44
7	Leaf apex shape	Obtuse	33.33	0.63
		Acute	66.67
8	Hair on the leaf	Present	16.67	0.45
		Absent	83.33
9	Leaf position	Opposite	100.00	0.00
10	Leaf size	Small	25.00	0.99
		Medium	55.55
		Large	19.44
11	Leaf density	Intermediate	22.22	0.53
		High	77.78
12	Leaf lobation	Shallow	86.11	0.40
		Deep	13.89
13	Waxiness on leaf	Absent	83.33	0.45
		Present	16.67
14	Tip color	Light green	63.89	0.85
		Dark green	27.78
		Purple	8.33
15	Petiole color	All green with purple base	13.89	1.64
		All green with purple leaf junction	8.33
		All purplish green with purple base	33.33
		All purplish green with purple leaf junction	11.11
		All purplish green with purple at both ends	25.00
		Green	8.33
16	Petiole wing color	Green	16.67	0.90
		Green with purple edge	63.89
		Purple	19.44
17	Twing direction	Clockwise	94.44	0.21
		Anticlockwise	5.56
18	Vine color	Green	41.67	0.68
		Light green	58.33
19	Wing on vine	Present	100.00	0.00
20	Wing color	Green	52.78	0.69
		Green with purple edge	47.22
21	Present/absent of wing	Present	30.55	0.61
		Absent	69.45
22	Spine shape	Straight	8.33	0.79
		Curved	22.22
		None	69.44
23	Color at spine base	Absent	69.44	0.61
		Present	30.56
24	Spine on vine	Few	25.00	0.96
		Many	16.67
		None	58.33
25	Tuber shape	Oval	8.33	1.39
		Oval oblong	11.11
		Cylindrical	36.11
		Flattened	8.33
		Irregular	36.11
26	Tuber flesh color	White	19.44	1.59
		Purple	19.44
		Purple with white	13.89
		White with purple	25.00
		Outer purple/ inner white	22.22
27	Tuber skin color	Light brown	50.00	0.69
		Dark brown	50.00
28	Tuber branching	Slightly branched	13.89	1.26
		Branched	44.44
		Highly branched	27.78
		None	13.89
29	Tuber surface texture	Smooth	22.22	0.53
		Rough	77.78
30	Hairiness of tuber surface	Small	22.22	0.53
		Medium	77.78
31	Flower	Present	25.00	0.56
		Absent	75.00
32	Flesh color at central transverse cross section	White	38.89	1.50
		Light purple	13.89
		Purple	13.89
		Purple with white	13.89
		White with purple	19.44
Overall Mean				0.72

 

Principal components analysis:

The patterns of variation and the relative importance of each trait in explaining the observed variability assessed through principal component analysis (PCA). In the present study, the principal component analysis was adopted based on 29 variables. The first seven principal components explained 88.4% of the total variation (Table 6). The first principal component (PCI) alone accounted for 32.5% of the total variation. Flesh color at central transverse, spine on vine base and tuber flesh color had the highest loadings on PCI. The second principal component (PC 2), explaining 22.8% of the total variation, was highly correlated with presence of spines on vine base and tuber flesh color, while PCIII associated with the spine on vine base, flesh color at central transverse, petiole and leaf color explained 12.9% of the total variation. The remaining PCs accounted 20.2% of the total variation, and mainly associated with petiole color, spine on vine base, tuber shape, flesh color at central transverse, tuber skin color and tuber branching (Table 6). From all the characters, tuber flesh color was found to be the most discriminative parameter differentiating landraces collected from southwest Ethiopia. To evaluate the scores of solitary landrace, PC1 and PC2 were plotted (Figure 3). All landraces distributed at the origin of the plot. The landrace 15/2000, 54/02 and 10/002 had the highest positive scores for both components and grouped to the top central corner of the plot. Conversely, landraces 75/02, 76/02, 27/02 and 59/02 had the lowest negative scores for PC-2 and the highest positive values for PC1 and grouped into the bottom right corner of the plot. The result of this study is consistent with the separation of landraces into two groups by UPGMA clustering (Figure 3).

 

 

Table 6. Eigen values, Proportion, Cumulative variance and component scores of the first seven principal components for qualitative traits in 36 yams from Southwest Ethiopia.

Variable	PC1	PC2	PC3	PC4	PC5	PC6	PC7
Eigen value	12.925	9.095	5.120	3.695	1.752	1.487	1.113
Proportion	0.325	0.228	0.129	0.093	0.044	0.037	0.028
Cumulative	0.325	0.553	0.682	0.775	0.819	0.856	0.884
Leaf color	-0.005	0.066	0.157	-0.037	0.148	-0.290	-0.113
Leaf vein color upper surface	0.037	0.032	-0.034	-0.128	-0.013	0.239	0.097
Leaf vein color lower surface	-0.013	0.029	-0.006	0.021	-0.009	0.018	-0.051
Leaf margin color	0.019	-0.024	-0.006	0.024	-0.134	-0.249	-0.466
Leaf shape	0.055	-0.075	-0.287	0.220	0.560	-0.535	0.060
Leaf apex shape	-0.043	-0.006	-0.038	-0.020	-0.071	0.123	0.105
Hair on leaf surface	0.010	-0.021	-0.019	-0.004	-0.027	0.010	-0.067
Leaf size	-0.003	-0.056	0.077	0.057	-0.091	0.128	-0.084
Leaf density	0.021	0.057	-0.034	-0.041	0.005	-0.023	-0.006
Leaf lobation	0.044	0.066	-0.012	0.021	-0.136	-0.043	0.014
Waxiness on leaf	0.008	0.050	0.007	0.057	0.105	-0.030	0.042
Tip color	0.046	0.066	-0.107	0.107	-0.039	-0.174	0.043
Petiole color	0.015	0.095	0.507	-0.710	0.220	-0.209	0.076
Petiole wing color	0.036	0.071	-0.049	0.091	0.024	-0.071	-0.247
Twining direction	-0.014	-0.006	-0.007	0.075	0.066	0.016	0.030
Vine color	-0.030	-0.045	0.018	0.002	0.010	0.063	0.129
Wing color	-0.029	0.011	-0.026	0.103	-0.101	-0.094	0.152
Presence/absence of spine	0.070	0.039	0.084	-0.008	0.081	0.108	-0.031
Spine shape	0.131	0.054	0.137	-0.016	0.124	0.164	-0.089
Color at spine base	0.062	0.022	0.078	0.027	0.123	0.100	-0.024
Spine on vine base	0.482	0.498	0.428	0.410	-0.165	-0.144	0.076
Tuber shape	0.008	0.062	-0.255	-0.269	-0.578	-0.458	0.338
Tuber flesh color	0.461	-0.812	0.258	0.094	-0.114	-0.103	0.083
Tuber skin color	0.020	-0.078	-0.051	-0.090	0.238	-0.071	0.119
Tuber branching	0.010	0.067	0.052	0.159	0.207	0.087	0.669
Tuber surface texture	-0.036	0.008	-0.042	0.010	-0.024	-0.013	-0.051
Hair on tuber surface	0.026	0.080	0.091	-0.040	-0.031	-0.213	-0.075
Flower	0.040	0.016	0.045	-0.016	0.073	-0.087	-0.096
Flesh color at central transverse	0.717	0.155	-0.496	-0.317	0.125	0.157	-0.061

 

 

Figure 3.The Bi-plot diagram of PCA I and PCA II of 36 yam landraces based on 29 qualitative traits.

 

 

CONCLUSION:

Analysis of Dioscorea spp. landraces based on phenotypic variability had a great contribution for better assessment of the landraces and identification of the plants with desired characteristics for breeding. Based on the result of the Shannon-Weaver diversity index, and cluster analysis revealed that maximum diversity existed between Dioscorea spp. landraces indicated the fact that hybridization among the landraces included with them would produce preeminent hybrids and desirable segregants. Besides, they appeared lack of reasonable degree of correspondence between groups of landraces and collected geographical location. To further investigate the genetic basis of the diversity revealed among Dioscorea spp. landraces, it is necessary to assess the existed landraces through additional collection from different growing areas over more seasons in different locations to get stable landraces for different environment.

 

REFERENCES:

1.      Afifi, A. and V. Clark .1990. Computer aided multivariate analysis. Van Nostr and Reinhold. New York, USA.

2.      Asiedu, R. and S. Alieu.2010. Crops that feed the world 1. Yams, Journal of food Science, 2: 305-315.

3.      Ayensu, E.S. and DG. Coursey. 1972. Guinea yams. The botany, ethno botany, use and possible future of yams in West Africa. Econ. Bot, 26: 301–318.

4.      Burkill, I.H. 1960. The organography and the evolution of the Dioscoreaceae, the family of the yams. J. Linnean Soc. (Botany), 56(367): 319–412.

5.      Cochran, W.G. and G.M. Cox .1957. Experimental designs.2^nd ed. John Wiley and Sons. New York.

6.      Coursey, D.G. 1967.Yams, an account of the nature, origin, cultivation and utilization of the useful members of the Dioscoreaceae. Tropical Agriculture Series. Longmans: London, greens and co Limmited, UK, 70-74.

7.      Dagne, Y. 2007.Studies on indigenous production and evaluation of landrace taro clones [Colocasia esculenta L. (Schott)] at Dalbo watershed, Wolaita, South Ethiopia. M.Sc. thesis, Presented to School of Graduate Studies, Hawassa University, Hawassa.

8.      Dansi A., H.D. Mignouna, R. Asiedu and F.M. Quin. 1999. Morphological diversity, cultivar groups and possible descent in the cultivated yams (Dioscorea cayenensis/D.rotundata complex) in Benin Republic. Genetic Resource and Crop Evolution, 46: 371-388.

9.      Dansi, A., B. Dantsey and R. Vodouhè. 2013. Production constraints and farmers’ cultivar preference criteria of cultivated yams (Dioscorea cayenensis /Dioscorea rotundata complex) in Togo, International Journal of Biology, 4:191-199.

10.   Dewey, D.R., and K.H. Lu. 1959. Correlation and path coefficient analysis of components of Crested Wheat grass seed production, Agronomy Journal, 51(9):515-518.

11.   Hahn, S.K and Y. Hozio. 1993. Sweet potato and yam. Symposium on potential productivity of field crops under different environments. Outlook on agriculture,1:10-11

12.   Hennink, S. and A.C. Zeven. 1991. The interpretation of Nei and Shannon-Weaver within population variation indices, Euphytica, 51: 235-240.

13.   Hunter, M .1996. Fundamentals of conservation biology. Blackwell Science, Inc., USA.

14.   IPGRI. 1997. Descriptors for yam (Dioscorea spp.). International Institute for Tropical Agriculture, Ibadan, Nigeria/International Plant Genetic Resources Institute, Rome, Italy.

15.   Jayasurya, A. 1984.Systematic arrangement of the genus Dioscorea (Dioscoreaceae) in Indian Sub-continent, Revised hand book to the Flora of Ceylon IX. Royal Botanic Gardens, Kew Richmond, UK.

16.   Karamura, DA. 1998. Cultivar distribution in primary banana growing regions of Uganda, MUSAFRICA, 9:3-5.

17.   Kifle, A. 2006. Characterization and divergence analysis of some Ethiopian taro (Colocasia esculenta (L.) accessions, M. Sc thesis, Alemaya University.

18.   Labouisse, J.P. 2006. Summary of passport data of coffee germplasm maintained at Jimma Agricultural Research Center (JARC), Un-published.

19.   Lebot, V. 2009. Tropical tuber and tuber crops: cassava, sweet potato, yams, aroids. CABI: Wallingford, Oxfordshire.

20.   Mignouna, H.D., A. Dansi and S. Zok.2002. Morphological and isozymic diversity of the cultivated yams (Dioscorea cayenensis /D. rotundata complex) of Cameroon. Genetic Resources and Crop Evolution, 49: 21-29.

21.   Milligan G. W and Cooper M.C. 1985.An examination of procedures for determining the number of cluster in data set. Psychometrika, 50(2):159-179.

22.   Mohammadi, S.A. and B.M. Prasanna.2003. Analysis of genetic diversity in crop plant Salient statistical tool and consideration. Crop science, 43: 1235-1248.

23.   Mulualem, T. and H. Mohammed.2013. Genetic diversity of Aerial Yam [Dioscorea bulbifera (L.)] accessions in Ethiopia based on agronomic traits. Science Publishing group, Journal of Agriculture, Forestry and Fishery, 2(2): 67-71.

24.   Mulualem, T. 2008. Morphological characterization and preliminary evaluation of aerial yam (Dioscorea bulbifera) accessions collected from southwest Ethiopia. M.Sc. thesis, Presented to School of Graduate Studies, Hawassa University, Ethiopia.

25.   Nebiyu, A. 2003. Characterization and divergence analysis for cassava [Mannihot esculenta Crantz (L.)] collection at Jimma, M.Sc thesis, Alemaya University, Ethiopia.

26.   Norman, P.E., P. Tongoona, J. Danson and P.E. Shanahan. 2012. Molecular characterization of some cultivated yam (Dioscorea spp.) genotypes in Sierra Leone using simple sequence repeats. International Journal of Agronomy Plant Production, 3(8): 265-273.

27.   Rhman, M.M. and A.Z.A. Munsur. 2009. Genetic divergence analysis of lemon. Journal of Bangladesh Agricultural University, 7(1):3337.

28.   SAS Institute. 2000. Statistical Analytical Systems SAS/STAT user’s guide version 8(2) cary NC: SAS institute inc.

29.   Sesay, L., P.E. Norman, A. Massaquoi, M.L Gboku and S.N. Fomba.2013. Assessment of farmers’ indigenous knowledge and selection criteria of yam in Sierra Leone. Sky Journal of Agricultural Research, 2(1):1–6.

30.   Silvia, B. 2006. Molecular characterization of Colombian yam by Selective Amplification of Micro satellite polymorphic Loci, 8(2):1-12.

31.   Smith, J.S.C. and O.S. Smith.1989. The description and assessment of distance between inbred lines of maize. The use of morphological traits as descriptors. Maydica, 34:141-150.

32.   Tamiru, M. 2006. Assessing diversity in yam (Dioscorea spp.) from Ethiopia based on morphology, AFLP marker and tuber quality, and farmers’ management of landraces. Ph.D. Thesis, George August University. Germany. Tamiru, M., C. Heiko, B. Becker and L. Maass. 2007. Genetic Diversity in Yam germplasm from Ethiopia and their relatedness to the main cultivated Dioscorea species assessed by AFLP markers. Journal of Crop Science, 47:1744–1753.

33.   Tamiru, M., C. Heiko, B. Becker and L. Maass. 2011. Comparative Analysis of Morphological and Farmers’ Cognitive Diversity inYam Landraces (Dioscorea spp.) from Southern Ethiopia, Tropical Agriculture Development, 55(1):28 – 43.

34.   Terauchi, R., V.A. Chikaleke, S. Thottappilly and K. Hahn. 1992. Origin and phylogeny of Guinea yams as revealed by RFLP analysis of chloroplast DNA and nuclear ribosomal DNA. Theor. Appl. Genet. 83: 743-751.

35.   Wilkin, P.1998. A morph-metric study of Dioscorea quartiniana A. Rich (Dioscoreaceae). Kew Bulletin 54:1–18.

36.   Zeven, AC. and J.M. De Wet.1982. Dictionary of cultivated plants and their region of diversity: Center for Agricultural Publication and Documentation (Wageningen).

 

 

 

 

Received on 26.02.2019         Modified on 20.03.2019

Accepted on 05.04.2019       ©A&V Publications All right reserved