Assessment of growth and yield of some high-and low-cyanide cassava genotypes in acid ultisols of south eastern Nigeria

Thirty-eight high and low cyanide cassava genotypes were evaluated for growth and yield. The experiment was fitted in a randomized complete block design with three replications. Analysis of variance (ANOVA) revealed significant difference (P<0.05) in growth and yield parameters, indicating strong genetic variations among the genotypes. Significant (P<0.05) highest average storage root weight of 1.5 and 1.8 kg were obtained from TMS 94/0035 (high cyanide cassava genotype) and TMS 98/0505 (low cyanide cassava genotype), respectively. Correlation analyses indicated significant (P<0.05) and positive associations between number of storage roots per plant, average storage root weight and root yield in the two types of cassava genotypes tested, indicating that the parameters are good indices that can be improved upon during breeding and selection to enhance root yield. Four high cyanide cassava genotypes (TMS 99/2123, TMS 96/1642, TMS 98/0068 and TMS 94/3200﴿ were outstanding in fresh root yield performance. To improve fresh root yield in cassava, the interrelationships among the various agronomic character of the cassava genotypes demands appropriate attention. The study gives more information on the improvement of fresh storage root yield of cassava.


INTRODUCTION
In the humid and sub-humid topics, cassava, which is one of the most important food crops play a dominant role in the rural economy and food security of the people living there.The root crop is characterized by the presence of cyanoglycosides (linamarin and lotoaustralin), which make both the roots and leaves potentially toxic to man and animals (Padmaja, 1995).Two distinct cassava genotypes exists namely, the bitter (high cyanide) and sweet (low cyanide) genotypes and their performance in terms of growth, photosynthetic efficiency and yield vary greatly.Santana et al. (2002) in their physiological studies on cassava reported that the content and quantity of cyanoglycosides in cassava roots depend on the genotypes and the growth conditions the plants are exposed to such as drought and soil composition.Eke-Okoro (2000) assessed the photosynthetic efficiency and productivity of low and high cyanide cassava genotypes and reported that high cyanide cassava genotypes had better photosynthetic efficiency and higher storage root yield than low cyanide cassava genotypes.Also, cassava genotypes with profuse branching characters have the tendency to produce higher fresh root yield than the less branching types (Eke-Okoro et al., 2001).
The breeding and release of cassava varieties by the *Corresponding author.E-mail: emmaukmbah@gmail.com.(Cock, 1986;Akoroda, 2005;and IITA, 1990).Furthermore, Githunguri et al. (2004) reported that cassava cultivated under humid conditions with high rainfalls have higher tuberous root bulking ability and lower cyanogenic potential than those cultivated in drier conditions.In recent years, studies have been directed at improving fresh storage root yield of cassava, one of which involves the analysis of inter-relationship among important agronomic characters in high-and lowcyanide cassava genotypes and some of them include the works of Asante and Dixon (2002), Akinwale et al. (2011) and Mulualem and Ayenew (2012).
The purpose of the study therefore was to assess the effect of genotypic difference of high and low cyanide cassava on root yield and other related characters, assess the performance of the genotypes as basis for ascertaining their acceptability and recommendation to farmers in the humid agro-ecozone of south-eastern Nigeria.Also, to clarify the inter-relationship between storage root yield and other agronomic characters of the genotypes as to elicit information that can be useful in boosting root yield of cassava during breeding and selection.

MATERIALS AND METHODS
Thirty-eight high-and low-cyanide cassava genotypes were obtained from the germplasm unit of National Root Crops Research Institute, Umudike, Nigeria and used for the study (Table 1).The experiment was carried out at Michael Okpara University of Agriculture, Umudike, Research Farm (latitude 05° 29 ' N, longitude 07 ° 33 ' E, altitude 122 m) in the low-land humid tropics of south eastern Nigeria in 2004 and 2005 cropping seasons.The treatments were laid out in a randomized complete block design (RCBD) with three replications.The total annual rainfall was 1,911.4mm (2004) and 2,064.8mm (2005).The soil, which is a sandy loam classified as ultisol (Paleulstalt), had, at 0 -20 cm depth a pH of 4.86 (1:2.5;Soil:Water), 1.62% organic matter; 0.56% total N; 26.0 mg/kg available P, which was high; 0.164 cmol/kg exchangeable K; and 0.34, 2.40 and 0.80 cmol/kg of Na, Ca and Mg, respectively.
The soil samples from the experimental plots were bulked, thoroughly mixed and then one composite sample obtained from which, a sub-sample was collected, air-dried and subjected to analyses in the laboratory.Soil pH was measured potentiametrically in a glass electrode in deionized water (pH water) at soil:water solution ratio of 1:2.5.Percentage organic matter was analysed using wet oxidation method by Walkley and Black (1934), while total nitrogen (N) was obtained by Microkjeldahl method of wet oxidation (Bremner, 1996).Available phosphorus was determined by calorimetric method (Olson and Sommers, 1982).Exchangeable potassium (K) and sodium (Na) were determined by Flamephoto-metory (Udo and Ogunwale, 1978), while calcium (Ca) and magnesium (Mg) were determined by ethylenediaminetetraacetate (EDTA) titration method (Olson and Sommers, 1982).
Twenty-five (25) cm long planting setts obtained from 12-month old matured cassava stems were planted at a spacing of 1 m apart on the crest of ridges spaced 1 m apart, which gave a plant population of 10,000 plants/ha.Each genotype occupied two rows of ridges measuring 20 m in length.A mixture of grammaxon (2 L/ha) and primextra (3 litres/ha) was applied to the field immediately after planting and two manual weeding regimes were carried out at 4 and 8 months after planting (MAP) to control weeds.N:P:K:Mg 12:12:17:2 fertilizer was applied at 1 MAP at the rate of 400 kg/ha.
Growth data were taken on plant height, number of stems per plant, number of internodes per plant and number of leaves per plant at 6 MAP.Yield data collected were on number of storage roots per plant, average storage root weight (kg) and fresh storage root yield (t/ha) at 12 MAP.Analysis of variance (ANOVA) was performed on growth and yield related traits following the procedure outlined for randomized complete block design (Steel et al., 1997).Data were also subjected to simple and partial correlation and regression analyses using SPSS statistical package for windows version 17.0 (2010).Mean separation was done using least significant difference (LSD) at 5% probability level according to Obi and Obi (2002).

RESULTS AND DISCUSSION
The ANOVA revealed that there was significant difference (P<0.05)among the cassava genotypes in all the plant characters assessed such as plant height, number of stems per plant, number of internodes per plant, number of leaves per plant, number of storage roots per plant, average storage root weight and fresh storage root yield (t/ha) (Table 2), indicating the presence of genetic differences.Plant height, number of stems per plant, number of internodes per plant and number of leaves per plant ranged between 1.20 and 2.30, 1.0 and 2.6, 195.0 and 444.0 as well as 52.0 and 288.7, respectively.
TMS 99/2123, TMS 98/0581, and TMS 97/4779 cassava genotypes were significantly (P<0.05)taller compared with the other genotypes except TMS 91/02324, TMS 96/1642, TMS 97/3200, TMS 96/1317, TMS 98/0068 and TMS 96/0603 while TMS 98/0581 significantly (P>0.05) had the lowest number of leaves per plant (52.0) relative to the other cassava genotypes.TMS 96/1569 had significantly highest number of internodes per plant (444.0)followed by TMS 97/3200 (438.7), which also had significantly highest number of leaves per plant, while TMS 96/1317 gave the highest number of stems per plant (2.6) relative to the other high cyanide cassava varieties.The results obtained corroborated studies by Okonkwo (2002) on evaluation of cassava genotypes for yield and biotic stress in which he reported significant differences in plant height and other growth parameters between cassava genotypes.
In high cyanide cassava genotypes, number of storages roots per plant, average storage root weight (kg) and fresh storage root yield (t/ha) ranged from 3.5 -6.5, 0.5-1.5 kg and 23.7 -63.4 t/ha, respectively.TMS 96/1642 gave the highest (P<0.05)number of storage roots per on evaluation of photosynthetic efficiency and productivity of sweet and bitter cassava varieties in the humid tropics as well as Okonkwo (2002) on some cassava genotypes in the cool highlands of Jos Plateau, Nigeria.Table 3 shows that there was significant difference (P<0.05)among the low cyanide cassava genotypes for plant height, number of stems per plant, number of internodes per plant, number of leaves per plant and average storage root weight.TMS 99/6012 significantly (P<0.05)exhibited the highest value for plant height (3.0 m) but had lowest number of leaves per plant (45.7) compared to the other genotypes.The differences in number of stems per plant and number of internodes per plant were significant (P<0.05).TMS 98/0505 produced the highest number of internodes per plant (548.0), which was higher by 65.7% relative to TMS 97/0325, which had the lowest number of internodes per plant (188.0), while TMS 4(2)1425 and TMS 92/0326 exhibited the highest values of 2.3 stems per plant.TMS TM 98/0028 had the highest number of leaves per plant while TMS 98/0505, TMS 92/0057 TMS 97/4769, and TMS 92/0326 were intermediate.The other genotypes exhibited the lowest values.These findings were in consonance with studies by Naskar et al. (1989) in which they surmised that differences in growth parameters could be basically due to high genetic variability among cassava genotypes, though environmental factors may also be considered.The number of storage roots per plant and fresh storage yield among low cyanide genotypes did not differ statistically (P>0.05).However, average storage root weight was significantly (P<0.05)highest in TMS 98/0505 (1.8 kg), followed by TMS 92 B /0068 (1.7 kg), while lowest values were obtained in TMS 97/4769 and TMS 82/0058 (0.7 kg).
The correlation between fresh storage root yield and the plant characters studied showed significant (P≤0.05) and positive relationship between number of storage roots per plant and fresh storage root yield with correlation coefficients (r) of (0.59) as well as between average storage root weight and storage root yield with (r = 0.58) in high cyanide cassava genotypes, an indication that these yield parameters are good indices that can be improved upon to boost cassava storage crop yield during breeding and selection (Table 4).The other parameters (plant height, number of stems per plant, number of internodes per plant and number of leaves per plant) did not show any correlated response with storage root yield of cassava.In low cyanide cassava genotypes, number of leaves per plant, number of internodes per plant and number of tubers per plant were significantly (P≤0.05)correlated with fresh storage root yield while average storage root weight had highly significant (P≤0.01)correlation with fresh storage root yield of cassava with (r = 0.74) (Table 5).These parameters demand close attention in breeding for improved storage root yield.Similar studies by Amadi et al. (2008) on potato genotypes showed that tuber number and average tuber weight are major individual contributors to crop yield.
Table 6 shows that analysing the plant characters using the method of partial correlation coefficients further highlighted the importance of average storage root weight and number of storage roots per plant with correlation   coefficients (partial r) of 0.81 and 0.73, respectively, for high cyanide cassava genotypes, as well as 0.87 and 0.76, respectively for low-cyanide cassava genotypes as vital characters that contributed greatly to storage root yield of cassava.Similar studies by Ntawuruhunga et al. (2001) on twenty broad-based cassava genotypes in the forest savanna, northern and southern guinea savanna as well as Sudan savanna agro-ecozones of Nigeria indi-cated increased storage root yield of cassava, which was mainly due to increase in number of storage roots per plant as well as individual storage root weight of the tested genotypes.This implies that for breeding and selection, premium should be placed on increasing the number of storage roots per plant and weight of average storage root in other to obtain higher root yield in both high-and low-cyanide cassava genotypes.

Conclusion
The results showed that among the thirty-eight high and low cyanide cassava genotype tested, four high cyanide cassava genotypes (TMS 99/2123, TMS 96/1642, TMS m 98/0068 and TMS 97/3200) were outstanding in fresh storage root yield performance, hence can be considered for recommendation to farmers in the humid agro-ecozone of southeastern Nigeria.Furthermore, the correlation studies specifically reviewed the inter-relationships among the various agronomic character of the cassava genotypes, which indicated that cassava selection programmes based on number of storage roots per plant and average weight of storage root, demands appropriate attention, especially when the challenge is aimed at improving fresh storage root yield in cassava.

Table 1 .
High-and low-cyanide cassava genotypes used in the study.
potentials.A number of researchers have stressed that tuber bulking and yield in cassava are determined by the characteristics of the genotype, assimilation supply of the plant and environmental factors

Table 2 .
Growth and yield characters of twenty five high-cyanide cassava genotypes in 2004 and 2005 cropping seasons.

Table 3 .
Growth and yield characters of thirteen low-cyanide cassava genotypes in 2004 and 2005 cropping seasons.

Table 4 .
Simple correlation matrix for the relationship between high-cyanide cassava genotype characters.

Table 5 .
Simple correlation matrix for the relationship between low-cyanide cassava genotype characters.

Table 6 .
Partial correlation coefficients between fresh storage root yield of cassava and six agronomic characters of high-and lowcyanide cassava genotypes.