 |
| About Bioline | All Journals | Testimonials | Membership | News |
|
||||||
|
||||||
African Crop Science Journal, Vol. 9, No. 1, March 2001, pp. 49-57
M. Mcharo, E. E. Carey1 and S. T. Gichuki2
Code Number: CS01034
ABSTRACT The major constraints to sweetpotato production in Kenya are use of low yielding albeit well adapted varieties and poor agronomic practices. Utilisation of higher yielding varieties could therefore increase yields. This study was conducted to evaluate introduced germplasm and identify clones with desirable traits. Twenty five clones selected on the basis of high root yield and high dry matter content were evaluated during the 1998 and 1999 long rains seasons. From these, 12 clones were selected for further evaluation in the 2000 long rains. Significant (P = 0.01) genotypic effects were observed for all traits, while genotype x season interactions were significant (P = 0.05) for large and total root yields. Twenty two of the clones had significantly lower foliage yields, compared to the check variety Mafuta. Clones 282-H-12 and WAGA-L-4 consistently recorded dry matter content of over 30%, which is above the critical value of 27%. Mean total root yields were highest in 1999 (14.6 tons ha-1) followed by 1998 (12.0 tons ha-1). The best performing clones were TZ-H-7, 120-H-11, 282-H-12 and TZ-H-10, all producing over 20 tons ha-1 with TZ-H-7 yielding significantly higher than the check variety Naveto. TZ-H-7 also recorded the second highest yield of 13.3 tons ha-1 during 2000. The consistently high number of clones outyielding the check varieties suggests a gain from introduction of new germplasm into the breeding programme in Kenya. Key Words: Dry matter, foliage production, germplasm, sweetpotato breeding RÉSUMÉ L'usage des variétés à bas rendement, malgré leur adaptation, et l'utilisation des méthodes culturales rudimentaires restent des contraintes majeures à la production de la patate douce au Kenya. L'utilisation des variétés à haute performance pourraient, cependant, améliorer les rendements. Cette étude a été conduite pour évaluer le germplasme introduit et identifier des clones avec des traits désirables. Vingt cinq clones choisis sur base de leur haut rendement des tubercules et du contenu en matière sèche de ces derniers, ont été évalués lors des longues saisons des pluies de 1998 et 1999. De ceux-ci, 12 clones furent retenus pour une évaluation plus poussée dans la longue saison des pluies de l'an 2000. Les effets génotypiques significatifs (P=0,01) furent observés pour tous les traits alors que des interactions génotype x saison furent significatives (P=0,05) pour le rendement total en tubercules et en tubercules de large taille. Parmi ces clones en évaluation, 22 eurent des rendements significativement faibles en feuillage comparés à la variété témoin, Mafuta. Les clones 282-H-12 et WAGA-L-4 enregistrèrent constamment le contenu en matière sèche au dessus de 30%, bien au dessus de la valeur critique de 27%. Les rendements moyens en tubercules furent les plus élevés pour l'année 1999 (14,6 t/ha) et ensuite pour l'année 1998 (12,0 t/ha). Les clones plus performants furent TZ-H-7, 120-H-11, 282-H-12 et TZ-H-10, et, tous produisirent plus de 20 t/ha, avec exceptionnellement TZ-H-7 produisant significativement plus que le témoin Naveto. TZ-H-7 enregistra aussi le second plus haut rendement de 13,3 t/ha pour la saison de l'année 2000. Le grand nombre de clones produisant plus que le témoin, d'une façon répétitive, suggère qu'il y at un gain dans l'introduction d'un nouveau germplasme dans le programme de sélection du Kenya. Mots Clés: Matière sèche, production du feuillage, germplasme, sélection de la patate douce INTRODUCTION Development of high yielding sweetpotato (Ipomoea batatas (L.) varieties for food and animal feed is a major goal of most sweetpotato breeding programmes. In Kenya this is even more important because a high percentage of sweetpotatoes is produced by small-scale farmers in densely populated high potential zones. In these zones the yields realised from farmers' fields are low due to land constraint, use of low yielding albeit well adapted varieties and poor agronomic practices. Increased productivity in such areas could therefore be achieved through application of appropriate agronomic practices and utilisation of higher yielding varieties (De Silva, 1995; Crossman, 1998; Gichuki et al., 1998). High yielding varieties can be identified by evaluation of existing landraces or introduction of exotic promising clones for further evaluation in formal trials. A number of past introductions into the Kenyan programme have given promising results and have been advanced for on-farm verification trials before release (Ndolo et al., 1998). The objective of this study was to make preliminary selections from the introduced germplasm and evaluate them on the basis of foliage and root yield and dry matter content. For preliminary evaluation and selection, the trials were conducted at a single site in Nairobi. Such focused procedure with set selection priorities assures positive progress early in the breeding programme (Jones et al., 1986). MATERIALS AND METHODS A total of 25 clones selected from polycross progeny, obtained from Uganda, on the basis of high root yield and high dry matter content were used in the first two trials in Nairobi. The parental clones are shown in Table 1. The 25 clones were evaluated in replicated trials during the long rains (March - July) of 1998 and 1999. Weather data during these periods are presented in Table 2. The experiments were laid out in a randomised complete block design with three replications. Each replication consisted of 30 plots, 25 of the plots consisted of the test clones while the other 5 consisted of check clones (Naveto, Maria Angola, KEMB 10, SPK 013 and Mafuta). Each plot had 44 plants planted in 4 rows, 11 plants per row. Between row spacing was 0.8m while within row spacing was 0.3m. Vine tip cuttings of 20 cm length were used as planting material and both trials were planted on flat. Gap filling was done two weeks after planting. Weeding was done until sufficient ground foliage cover to smother the weeds was achieved. Earthing-up was also done during weeding to seal any soil cracks through which roots could be exposed. The trials were harvested 5 months after planting. From the two middle rows in each plot, data were recorded on fresh foliage weight, large root weight, small root weight, total root weight, and the dry matter weights of the roots. Large roots had a diameter equal to or greater than 2.50 cm while small roots had a diameter less than 2.50 cm. From all measurements the corresponding yields in tons ha-1 were determined. Dry matter content was determined within 24 hours of harvesting. The middle portion of the fresh roots was cut into thin slices and then placed in open trays and dried in an oven at 60°C for 72 hours or until a constant dry weight was achieved. This weight was then recorded and from it % dry matter content was determined. Dry matter yield was a product of dry matter content and total root yield. Analyses of variance for the individual trials and for the combined trials were done according to Steel and Torrie (1980) and Gomez and Gomez (1984) using the MSTAT-C (1993) statistical package. Mean separation was done using LSD. Based on the results of the two trials a selection intensity of 50% was applied to select the 12 most promising clones on the basis of root yield and dry matter content. The 12 selected clones were evaluated in a further trial during the long rains of 2000. Experimental set-up, crop management practices, data collection and analyses were as in the earlier (1998 and 1999) trials. RESULTS AND DISCUSSION Results from the combined analysis of variance for 1998 and 1999 (Table 3) indicated significant seasonal effects on small root yield, total root yield and dry matter yield (P=0.01). Similar results have been reported by Ngeve (1993). Genotypic differences were significant for all traits (P=0.01) while Genotype x season interaction was only significant for large root yield and total root yield (P=0.05). Nawale and Salvi (1983) and Phemba et al. (1998) reported similar results for total root yield. During the 2000 trial season genotypic effects were significant for all traits measured (Table 4) at P=0.01. Data for the various traits are presented in Tables 5-7. Foliage yield. The yields of the tested clones ranged from 29.7 tons ha-1 (TOR-H-14) which was lower than for Mafuta (34.9 tons ha-1) to 8.7 tons ha-1 (440041-L-16) (Table 5). The mean yield for the third season (Table 7) was 19.9 tons ha-1. Small root yield. Mean small root yield contributed 13.5% to the mean total root yield, but 26% to the total root yield in the clones WAGA-L-4 and 277-H-1 (Table 5). The ratio of small root yield to total root yield remained constant during the 2000 trial season. However 33% of the total root yield for 69-H-1 consisted of small roots during the 2000 season (Table 7). A high percentage of small root yields suggests a longer period to maturity hence these clones may be useful for piecemeal harvesting or for production where soil moisture is not a constraint. Small roots are also unmarketable. Other clones that had proportionally high small root yield were NK-H-5, NK-H-7 and 400004-L-5 (Table 5), however, the total root yields of these three clones were too low for them to be considered for further evaluation during the 2000 season. Dry matter content and yield. Although NK-H-7 recorded the highest dry matter content (34.2%) it had a low total root yield (4.8 tons ha-1) and hence its high dry matter content is not of economic value. The standard checks for acceptable dry matter content in Kabete were KEMB 10 (30.8%) and SPK 013 (31.1%). Clones with a dry matter content of above 30% gave low yields with the exception of 282-H-12 (31.6%) and NK-L-22 (30.9%) (Table 5). These two clones also gave the highest dry matter yield. There was no discernible trend in seasonal influence on dry matter content (Tables 3, 5 & 7). The mean dry matter content during the 2000 trial (27.9%) was less than the combined mean recorded during the previous two seasons (28.9%). This could be due to the fact that a number of the initial clones, which were not ultimately selected, had high dry matter content. However, 282-H-12 and WAGA-L-4 consistently recorded dry matter content of over 30%. High dry matter content is a positive attribute in the Kenyan fresh market. Taste panels in Kenya have found dry matter content above 27% to be acceptable to most consumers (Carey et al., 1997). The watery texture of the low dry matter clones could be acceptable to infants, who may find its low dry matter content easier to digest than the drier textured varieties preferred by adults. Clones with a low dry matter but high yields may be used for processed products like flour production. In addition, in the drier areas, dry matter content may not be as critical as total root yield. Large root yield. Large roots are marketable roots, hence for human consumption clones with a high proportion of large root yields are desirable (Ngeve, 1991). Higher mean large root yield was realised in 1999 (12.1 t ha-1) than in 1998 (11.0 t ha-1) (Table 6). From the combined means the highest mean root yield was produced by TZ-H-7 (19.4 tons ha-1) followed by 120-H-11 (18.6 tons ha-1) and TZ-H-1 (18.1 tons ha-1). Two of these clones were selections originating from the widely cultivated parental variety Tanzania (TZ). The highest yielding check variety was Naveto (14.2 tons ha-1). Overall, nine of the clones gave higher mean yields than Naveto. The drought stress conditions that prevailed during the first two trial seasons (1998 and 1999) could have depressed yields of these clones in those seasons. In 1998 the trial received 194.7mm while in 1999 the trial received only 123.3mm (Table 2). Ekanayake et al. (1988) suggested that water stress reduces root yield in sweetpotatoes. Although the 2000 trial season received 270.2 mm (source: University of Nairobi meteorological data) the mean yield recorded (7.2 tons ha-1) was the lowest among the three trials. During this season an attack by the clearwing moth (Synanthedon spp.) was observed in all the plots and most likely was responsible for the depressed yields. Ames et al (1996) described in details damage by this pest. The lowest yielders for the combined 1998 and 1999 seasons were 400004-L-5 (1.8 tons ha-1) and NK-H-7 (3.5 tons ha-1). Clones 29-H-23 (13.4 tons ha-1), TZ-H-7 (11.6 tons ha-1) and 282-H-12 (10.5 tons ha-1) yielded significantly higher than Naveto (5.72 tons ha-1). Eight test clones gave higher yields than either KEMB 10 or Naveto. This suggests that there was a benefit from introduction of exotic germplasm into the breeding programme. In a similar study involving local and introduced open-pollinated progenies, Anselmo et al., (1998) reported higher yields from the introduced progenies than from the local clones. Total root yield. The mean total root yields were significantly (P=0.01) influenced by seasonal effects, the long rains of 1999 giving a higher mean yield (14.6 tons ha-1) than in 1998 (12.0 tons ha-1) (Table 6) and in 2000 (8.3 tons ha-1) (Table 7). Naskar and Singh (1992) and David et al. (1998) reported similar results. Total root yield ranged from 21.4 tons ha-1 (TZ-H-7) to 3.3 tons ha-1 (400004-L-5) among the test clones in 1998 and 1999 (Table 6). Overall, Mafuta was the lowest yielding clone (2.2 tons ha-1). Clones 120-H-12 and TZ-H-10 also produced over 20 tons ha-1. The highest yielding check clone was Naveto (15.6 tons ha-1). The results showed that TZ-H-7 and NK-L-1 yielded higher foliage and significantly (P=0.05) higher total root yield than Maria Angola. Maria Angola, a dual-purpose variety, recorded a total root yield of 12.6 tons ha-1. These clones therefore are promising for dual-purpose utilisation. TZ-H-7 and NK-L-1 also gave higher yields than Naveto, another dual-purpose variety both for foliage and root production. The NK-H selections, namely, NK-H-5 and NK-H-7 performed poorly, yielding 6.0 and 4.8 t ha-1, respectively. During the year 2000, eight clones yielded higher than the check varieties KEMB 10 and Naveto (Table 7). Clones 29-H-23 (14.6 tons ha-1) and TZ-H-7 (13.3 tons ha-1) gave the highest yields. The top four test clones had significantly higher yields than both KEMB 10 and Naveto. Gichuki et al. (1998) also reported higher yields from introduced progenies compared to local checks. CONCLUSION Results from these trials suggest a gain from the introduction of new germplasm
into the breeding programme. The results indicate significant environmental
effects on roots and foliage yield. The dry conditions under which the trials
were conducted, which were less than optimum for crop growth, provided good
conditions for screening for drought tolerance. ACKNOWLEDGEMENT We wish to thank the University of Nairobi for providing land and other facilities, Mr. Joseph Kitonyi for field management of the trials, and the International Potato Centre for funding the study. REFERENCES
TABLE 2. Weather data for 1998 and 1999 at Nairobi, Kenya
TABLE 3. Mean squares for yield traits
NS = Not significant
TABLE 4. Mean squares for yield traits during 2000
NS = Not significant
*, ** = Significant at 0.05 and 0.01 level of probability, respectively TABLE 5. Combined mean performance of 25 sweet potato clones during 1998 and 1999 long rains
*Large roots have a diameter >=2.5cm; small roots have a
diameter <2.5cm +Selected clones showing their positions relative to the
rest of the test clones
TABLE 6. Performance of sweetpotato clones in a preliminary yield trial planted in Nairobi during the long rains of 1998
TABLE 7. Performance of the selected sweetpotato clones during the 2000 long rains
Large roots have a diameter >=2.5 cm; small roots have a
diameter <2.5 cm
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| |||||||||
