The Bananas of Upland East Africa - PRINCIPAL ELEMENTS OF THE 1987-1990 RESEARCH PROGRAMME |
The foreign strategy for rehabilitation of the
banana shambas relied on certain assumptions which were untested locally.
For instance, definitive evidence was not available on the problems
actually faced by the crop. The recommendation of use of manure to improve
plant nutrition was unquantified. Similar uncertainty applied to both the
requirement for a break period and to its duration. The appropriateness of
the particular pesticide carbofuran also had not been extensively field
tested. Thus, the research programme targetted several basic elements.
These were:
In recent years, thinking on crop production has broadened considerably and it has become accepted that it is the cropping system as a whole that needs to be examined rather than a single, albeit visibly obvious, problem such as a pest. For instance, ZADOKS (1981) gave a schematic diagram outlining five levels of crop yield; primitive, actual, economic, attainable and theorical; and pointed out that yield is limited by production constraints of which "harmful organisms" is only one. He added that much work has yet to be done, especially for tropical crops. In 1983, KIRKBY emphasised that, despite the importance of bananas as a staple food crop for some 15 million people, a bibliography on crop production research in East Africa listed only six works on bananas out of 1716 references (R.A. KIRKBY, Discussion paper in Kirkby, R.A. & NGENDAHAYO, D. (Eds) Banana production and research in Eastern and Central Africa. Manuscript Report, IDRC-MR114e -132 pp Ottawa, Int. Dev. Res. Centre).
The limiting factor was defined by THOMASON & CRESWELL (1987) as that biotic or abiotic component of the system that is constraining the utilisation of inputs by the crop. Their definition, however, seems to presuppose that there is adequate provision of inputs and could be said to be starting at Zadok's "actual level" of crop yield. In Kagera, as in the other parts of the great upland banana growing area, much of the production hardly passes the "primitive level". The long-term permanent upgrading of banana yields will be achieved only when the limiting factor in the broader concept of including the clarification of the essential inputs has been identified. MBWANA (op cit., 1987) lists a number of possible contributory causes for the decline of the banana crop in Tanzania; apart from pests and diseases, weeds, poor husbandry and low soil fertility are cited, although no data are given in his report. Additionally, different banana growing areas may have different limiting factors and, therefore, require specific proposals for crop improvement. The ratio of costs to benefits also has to be determined for any such proposals.
THOMASON & CRESWELL (1987) cited the definition of a "key pest" as being an organism that causes significant reduction in crop yield every season unless some pest control action is taken. They further stated that management systems are built around key pests.
Banana weevils - As mentioned earlier, Cosmopolites sordidus has long been regarded as the key pest of bananas in the upland area of East Africa; as recently as 1987, DDUNGU, writing on regional needs for banana improvement in East Africa, stated that "this is a very serious pest in every country of the region". Despite such dogmatic statements, the fact is that none of the published reports give any sound data on actual crop loss caused by this insect. The most easily observed effect of C. sordidus on yield is the total loss of the bunch when a pseudostem weakened by larval tunnelling snaps at or just above ground level. This may be what BUJULO et al. (1986) referred to as "blow downs" but their text was not clear on this key issue. The danger, therefore is that it may well be that the effect of C. sordidus on the crop has been overestimated.
Some 85 years ago, JEPSON (1914) described the death of young suckers attacked by the borer but then went on to note that plants attacked at a later stage in their growth are often capable of producing a good crop, the presence of the borer not then seeming to affect them. VILLARDEBO (1973) described the female as showing a very accentuated preference for a rhizome which has attained or passed the flowering stage as the site of oviposition. He added that in a banana plantation with a moderate infestation of weevils the immature suckers are perfectly healthy even when one can find larval tunnels in the corms of fruiting plants or in the corms of harvested plants. There are recognised systems, developed elsewhere in the world, for estimating the degree of attack by banana weevil, such as the percentage coefficient of infestation methods described by VILLARDEBO (1973) and MITCHELL (1978) but these do not permit more than an external examination of corms for evidence of larval tunnelling. Peripheral damage to the corm so observed need not necessarily lead to disruption of nutrient conduction or of stem growth as these processes both occur in the central part of the corm. Thus, there was a clear need for an effort to develop a technique for testing the correlation, if any, between the incidence of banana weevil adults, the amount of larval attack and the actual crop loss.
Nematode pests - From the introduction, it can be seen that one or more nematode species may now have greater claim to being the "key pest" but again the evidence is of pest incidence and not of proven crop loss. Similarly to the C. sordidus situation, total loss of the bunch can occur when nematode infestation leads to severe root death and the plant is toppled, characteristically with the corm coming right out the soil. The degree of crop loss with lesser infestations or smaller less vigorous plants had not been established. Additionally, although there was a good amount of information available on the species Radopholus similis, WALKER et al. (op cit, 1983) concluded that in Kagera another species, Pratylenchus goodeyi, was of the greatest importance. There was almost no background information on P. goodeyi and, thus, key questions on its bionomics needed answering before a sound control strategy could be formulated. One needed to know, for instance; has P. goodeyi any alternative hosts; what is the survival of P. goodeyi in the soil in the absence of a host plant; and, how susceptible is P. goodeyi to a nematicide? To refer back to the "limiting factor" concept, there was evidence that soil amendments, such as the application of fertilisers or the incorporation of organic matter, can influence nematode populations either directly or indirectly (BRIDGE, 1987; BROWN, 1987). This was yet to be studied in the case of bananas.
NEXT
| ©2000 - Brian Taylor CBiol FIBiol
FRES 11, Grazingfield, Wilford, Nottingham, NG11 7FN, U.K. Visiting Academic in the Department of Life Science, University of Nottingham |
href="\BANANAS2000\RESPROPS.HTM"