+Bioline International Official Site (site up-dated regularly)

Variability in severity of potato late blight and yield loss associated with the disease was studied in the major potato growing agro-ecologies of Uganda. Late blight incidence and severity varied significantly (P<0.05) from district to district to season during the four surveys (2 years). Overall, the highest late blight severities were recorded in Kabale district (mean severity of 27.6%), followed by Mbale (21.8%), and Mbarara the least (8.7%). There was more disease during the month of December 1997, with mean severity of 40.2%, while the least severity was in May 1998 (10.8%). The highest severity was recorded on cultivar Victoria (62.1%), followed by Wafrika (61.8%), Sangema (59.5%) and Maboni (48.2%). Rutuku and Musitamya recorded the least severities, less than 20% during 1998/99. Overall, there was no significant correlation between late blight severity and yield of potatoes, likely because the yield loss assessment was done during a season of low late blight occurrence (season was dry).

Key Words: Phytophthora infestans, Solanum tuberosum, varietal resistance, Uganda

La variation de la sévérité du mildiou de la pomme de terre et les pertes de rendement associées à cette maladie ont été étudiée dans les régions agro-écologiques productrices de la pomme de terre en Uganda. L' incidence et la sévérité du mildiou ont varié significativement d'un district à l'autre et d'une saison à l'autre pour les quatre enquêtes conduites pendant les deux ans. Dans l'ensemble, les sévérités les plus élevées du mildiou dans le district de Kabale ont constamment montré une moyenne de 27.6%, suivies par Mbale (21.8%) alors que Mbarara venait en dernière position (8.7%). Il y avait plus de maladie pendant le mois de décembre 1997, avec une moyenne de 40.2%, alors que la faible sévérité était observée au cours du mois de mai 1998 (10.8%). La sévérité la plus élevée a été observée sur le cultivar Victoria (62.1%), suivi de Wafrika (61.8%), Sangema (59.5%), et Maboni (48.2%). Rutuku et Musitamya ont montré des moindres sévérités, moins de 20% pendant la saison 1998/99. Générallement, il n'y avait pas de corrélation significative entre la sévérité du mildiou et le rendement de la pomme de terre, probablement parce que l' évaluation des pertes de rendements a été faite pendant la saison où l' apparition de la maladie était faible ( saison sèche).

Mots Clés: Phytophytora infestans, Solanum tuberosum, résistance variétale, Ouganda

Potato (Solanum tuberosum L.) is an important food staple and cash crop in Uganda. It is grown predominantly in highland areas of western Uganda (Kabale, Kisoro, Mbarara, Kasese, Kabarole), eastern (Mbale, Kapchorwa) and north-western in Nebbi (Sikka et al., 1991). However, intensive potato cultivation in these areas is handicapped by several factors of which late blight (Phytophthora infestans (Mont.) de Bary), bacterial wilt (Ralstonia solanacearum E.M. Smith) and limited availability of improved seed are the most important. In Uganda, late blight is endemic and is destructive in highland areas, although occasional epidemics occur in lowland areas (Bhagsari et al., 1994). However, information on the actual variability in severity of late blight and yield loss attributable to the disease on potato in Uganda have not been qunatified. The objective of our study was to establish the incidence, severity and damage of late blight in two major potato growing agroecologies of Uganda.

Survey. Field surveys were conducted to establish the incidence, severity and damage of late blight on potato. Since diseases may show seasonal variability which can be missed in one-time surveys, four separate surveys were done, in December 1997, May 1998, November 1998 and May 1999. The survey was conducted in south western Uganda, i.e., the districts of Kisoro, Kabale and Mbarara, and eastern Uganda (Mbale) the major potato growing areas of the country (Sikka et al., 1991).
In each district, the number of potato fields sampled varied from 10 to 35. A total of 102 fields were surveyed in December 1997, 62 in May 1998, 88 in November 1998 and 80 in May 1999. Sampling of the crop was done 1 to 2 weeks after flowering (45 to 65 days old), the stage at which the crop tends to be most vulnerable to late blight epidemics (Turkensteen, 1988). Potato fields were sampled every 10 to 15 km of a road system across each district. The size of potato fields ranged between 0.2 to 2 hectares . When potato fields were seen, even outside the systematic interval, sampling would be made depending on the land size, crop population, stage of growth and topography. Incidence of late blight was obtained by counting the number of infected plants in a sample of 20 plants from each of the four different positions of each potato field, making a total of 80 plants per plot. The positions were selected randomly excluding borders. Each plant sampled was assessed for late blight severity basing on the 0-100% scale (Henfling, 1987). Information on type of fungicide used, number of sprays, and names of cultivars were also noted.

Sampling in the different seasons was done in the same villages, although not necessarily from the same farmers. Seasons and districts were considered as treatments and fields within each district were the replications. Incidence and severity data for all seasons were transformed using arcsine transformation (x+1), to harmonise variance and pooled together prior to Analysis of variance (ANOVA) using general linear model (GLM) procedure of the SAS statistical package (SAS, 1987). Means (transformed) were separated by Least Square means.

Effect of late blight on yield. Effect of late blight on potato yield was assessed only in farmers' fields in the districts of Kabale and Mbarara during the second season of 1998. Twelve potato fields (plots) were sampled during the study, 6 farmers from Kabale district, and 6 were from Mwizi sub-county in Mbarara district. The fields were assessed three times, during the second week of November 1998 (vegetative stage), during the first week of December 1998 (flowering stage) and the first week of January 1999 (mature stage). During each visit, severity of late blight disease was recorded from 3 randomly selected positions within each potato field as described earlier. On average, each sampling position measured 5 m x 1.4 m and were considered replications for data collection. At harvest the same sampling procedure was used. The data were subjected to Analysis of variance (ANOVA) using general linear model (GLM) procedure. Additionally, the relationship of late blight severities to tuber yield were analysed using correlation and regression methods. All statistical analyses were done using SAS Statistical Package (SAS, 1987).

Different districts were found to grow different potato varieties and some varieties were more common than others. Overall, Victoria was the most commonly grown variety in all the four districts, whereas Musitamya was common in the three districts except Mbale. Kabale district had the highest number of varieties grown (10) followed by Kisoro (7) , whereas Mbarara had only two varieties. Varieties Maboni and Wafurika where found only in Mbale (Table 1).

The districts surveyed had different cropping systems. In Kabale and Mbale, potatoes were mostly grown as a monocrop. In Mbale, the farming system in the area was predominantly a coffee-banana system, but on a few occasions, potato was intercropped with bananas or coffee. In Mbarara, potatoes were found commonly intercropped with beans (Phaseolus vulgaris L.) or field peas (Pisum sativa L.) and on a few occasions with maize (Zea mays L). In Kisoro, potatoes were mixed cropped with field beans, maize and sorghum (Sorghum bicolor), and occasionally, in pure stands. In all the districts surveyed, except Mbarara, pure stands (monocrops) recorded the highest amount of late blight with mean severity of more than 20% (Table 2). In Kisoro, potato+bean intercrop had a severity of 20%, and the potato+maize or potato+sorghum registered 18.7%. In Mbarara, potato+maize intercrop had the highest severity (10.2%). In the case of Mbale potato+banana intercrop had 23.6% severity and the least disease was from the potato+coffee cropping system (10.9%).

Late blight incidence and severity varied significantly (P≤0.05) from district to district and season to season during the four survey periods. Additionally, location (district) x season interaction was highly significant. During the first survey in December 1997, Mbale district recorded the highest amount of disease (99.1% incidence with 58.8% severity), Kabale recorded the least (82.6% incidence and 22.9% severity (Table 3). During the second survey in May 1998, disease severity in all the surveyed districts was lower than 20 % with Kabale recording the highest (16.6%). In November 1998 and May 1999, Kabale district still had the highest amount of disease with severities of 39% and 31.8%, respectively.

Overall, Kabale district consistently recorded the highest late blight severities, with mean severity of 27.6%, followed by Mbale (21.8%) and Mbarara consistently recorded the least, with mean severity of only 10.8%. Considering the four seasons, there was more disease during December 1997, with mean severity of 40.2%, while the least disease was in May 1998 (10.8%). The highest blight severity was recorded on the variety Victoria (62.1%) followed by Wafrika (61.8%), Sangema (59.5%) and Maboni (48.2%). In all the districts, Victoria, Sangema and Kabale had consistently higher late blight severities, while Musitamya and Nakpot 1 (382171.4) had the least (Table 1).

During the survey, it was noted that potato farmers used fungicides to control late blight. Dithane M-45 (Mancozeb 80% WP) was the most commonly used fungicide in all the four districts. Occasionally, Ridomil (systemic fungicide) was used in Kisoro, but more commonly in Kabale district. Where Ridomil was used, farmers applied it in combination with Dithane M-45 (contact fungicide). In addition to the two fungicides farmers in Kisoro and Mbale used copper compounds. The frequencies of fungicide use was higher in Mbale district, with a mean of five sprays per season. In Kisoro and Kabale farmers applied on average 3 sprays per crop, but those in Mbarara district applied only one spray per crop.

The effect of potato late blight on tuber yield was assessed only in Kabale and Mbarara districts during the second season of 1998. Late blight severity and tuber yield varied significantly (P≤0.05) within fields sampled, and also between the two districts. Additionally, location (district) x potato field interaction was significant (Table 3). Disease levels varied from variety to variety, with Victoria registering the highest amount of disease (48.3% severity), and Musitamya the least (5% severity). Other varieties had mean severies ranging between 5 to about 38.3% (Table 4). Likewise, average tuber yield varied from variety to variety with Sangema recording the highest (16.9 t ha^-1) and Musitamya the least (8.2 t ha^-1). Tuber yields averaged 14.6 and 7.1 t ha^-1 for Kabale and Mbarara plots, respectively. There was generally a weak negative correlation between late blight severity and yield of potatoes in the two districts (r=-0.065, P=0.71). The correlation coefficients between late blight severities and tuber yield for each of the varieties Cruza, Musitamya, Rutuku, Sangema and Victoria were 0.948, -0.39, 0.94, -0.960 and -0.980 at P=0.67, respectively.

Variation in late blight incidence and severity among the four districts was probably related to dry conditions and varietal differences in relation to late blight. Except for Mbale district, which had the greatest amount of disease in December 1997, there was generally more late blight in Kabale district. In 1997, Mbale district recorded rainfall above 200 mm per month just before the survey, this must have induced high late blight attack.

Varieties Victoria, Sangema and Kabale consistently recorded higher severities of late blight disease in all the districts and seasons. These varieties posses the R-genes (Umaerus et al., 1983), which may have succumbed to P. infestans pathotypes over the years. This calls for continued effort to develop potato varieties with stable (durable) resistance to P. infestans populations (Deahl et al., 1995; Goodwin et al., 1996). In Uganda, there are on-going efforts to develop potato varieties with durable resistance, and some clones within population B genotypes which lack R-genes have shown good promise (Hakiza et al., 1999).

The higher levels of late blight on the common varieties may also partially be due to fungicide resistance and improper use of fungicides (poor timing, low dosage, etc). Lack of late blight control has been attributed to the appearance of metalaxyl resistant strains of P. infestans (Deahl et al., 1995; Goodwin et al., 1996). In this study a number of farmers were found to use fungicides to control late blight on potatoes, Mbale farmers recording the highest frequency of application. Although there was more use of fungicides in Mbale than other districts, late blight was still severe in Mbale. In Kabale and Kisoro districts, there was evidence of use of the systemic fungicide ridomil. Studies elsewhere eg. in Netherlands and Ireland (Davidse et al., 1981; Dowley and O'Sullivan, 1981) have reported the development of P. infestans phenotypes resistant to metalaxyl, which were detected after the use of phenylamide compounds to which metalaxyl, an ingredient of ridomil belongs. In our other studies (Mukalazi et al., 2001) we detected high frequencies of metalaxyl resistant P. infestans strains in Kabale and Mbale samples.
Potatoes grown as pure stands (monocrops) recorded the highest amount of disease in all the districts. The high disease severity could be associated with the fact that, when potatoes are grown alone, they tend to form a canopy on establishment, thus creating a condusive microclimate for late blight development. In Kisoro, potato+bean intercrop also recorded high late blight severity. In this district, farmers commonly grow climbing beans, such that when grown in an intercrop with potatoes, this probably creates conducive environment for late blight development. Contrastingly, there was less late blight when potato was intercropped with bush type of beans as was the case in Mbale and Mbarara.

Overall, there was low association between late blight severity and yield of potatoes. There were two reasons for this. First, most of the farmers had applied fungicides to control late blight, hence the low disease levels. Also late blight yield loss study was only conducted during October-December 1998, a season which was exceptionally dry and therefore, only mild late blight developed. Mbarara recorded an average of 40 mm of rain while Kabale had 179.3 mm during the entire growing season.
The average yields of 14.6 and 7.1 t ha^-1 for Kabale and Mbarara, respectively, are quite low compared to on-station yields (26 t ha^-1) at Kalengyerein south western Uganda. The low values may be associated with low soil fertility since none of the farmers interviewed had applied organic or inorganic fertilizer in the fields sampled. On the otherhand the relatively high yields in Kabale compared to Mbarara is related to the more fertile volcanic soils in the district, and accessibility to high quality seed from Kalengyre Research Station.

Funding of this study was by the Global Initiative on Late Blight Project (GILB) through the International Potato Centre (CIP) with supplementary funding from the Rockefeller Foundation ACA grant to the second author.

TABLE 1. Mean late blight severities (%) on potato varieties in selected districts of Uganda during 1997- 1999

Varieties	Dec. 1997	May 1998	Nov. 1998	May 1999	Mean
Kisoro district
Cruza	42.7	12.4	16.1	4.1	18.8
Kabale	20.0	-	41.9	-	31.0
Mbumbamagara	-	-	-	20.4	20.4
Musitamya	-	-	11.6	-	11.6
Rutuku	38.4	10.4	18.4	5.7	18.2
Sangema	45.0	-	-	5.0	25.0
Victoria	46.2	5.8	24.7	4.5	20.3
Mean	38.5	9.5	22.5	7.9	20.7
Kabale district
Cruza	-	-	40.7	18.9	29.8
Kabale	26.2	18.6	65.7	-	36.6
Kisoro	-	-	49.5	58.3	53.9
Malirahinda	-	6.6	42.2	6.3	18.4
Mbumbamagara	-	12.1	22.5	50.0	28.2
Musitamya	-	-	-	0.0	0.0
Rutuku	-	15.4	32.1	46.1	31.2
Sangema	31.7	15.7	53.5	14.6	28.9
Victoria	12.8	14.8	42.4	36.6	26.7
Nakpot 1 (382171.4)	-	-	-	0.0	0.0
Mean	23.6	13.7	43.6	25.6	25.4
Mbarara district
Musitamya	*	-	0.1	0.3	0.2
Victoria	*	5.7	10.9	8.2	0.3
Mean	5.7	5.5	4.3	4.3
Mbale district
Maboni	48.2	0.5	0.3	16.3	16.3
Sangema	59.5	27.0	-	-	43.3
Victoria	62.1	0.5	-	-	31.3
Wafurika	61.8	8.8	-	20.5	30.3
Mean	57.9	9.2	3.8	18.4	25.5

*Survey not conducted during the season; Late blight assessment scale, 0-100% (Henfling, 1987)

TABLE 2. Influence of cropping system on late blight incidence and severity in Uganda, 1997-1999

Cropping system	Incidence (%)	Severity (%)
Kisoro district
Potato+beans	71.0	20.0
Potato+maize or+sorghum	70.0	18.7
Pure stand (monocrop)	75.0	20.4
Mean	72.0	19.6
Kabale district
Pure stand	85.9	28.3
Mbarara district
Potato+beans or +peas	44.9	8.2
Potato+maize	42.0	10.2
Pure stand	30.1	7.9
Mean	40.0	8.8
Mbale district
Potato+banana	49.2	23.6
Potato+coffee	42.0	10.9
Potato+peas	46.0	19.2
Pure stand	55.4	29.5
Mean	48.1	20.8

Pooled mean incidence and severities on different potato varieties during four cropping seasons

TABLE 3. Incidence (%) and severity(%) of potato late blight in different districts of Uganda, 1997-1999²

Season	Kisoro	Kabale	Mbarara	Mbale	Mean¹
December 1997
Incidence	88.3 abc	82.6 cd	-	99.3 a	90.1
Severity	38.7 b	22.9 c	-	58.8 a	40.2
May 1998
Incidence	73.9 cd	90.2 abc	65.9 de	25.3 f	63.8
Severity	12.6 cde	16.6 cd	10.0 def	4.1 ef	10.8
November 1998
Incidence	85.4 bc	95.2 ab	23.6 fg	5.0 f	52.3
Severity	20.4 c	39.0 b	9.7 def	4.5 f	18.4
May 1999
Incidence	52.1 e	81.6 bc	27.9 f	66.9 de	57.1
Severity	7.1 def	31.8 b	6.4 ef	19.7 c	16.3
Mean³
Incidence	74.9	87.4	39.1	49.1	62.6
Severity	19.7	27.6	8.7	21.8	19.5

1 Means in each column and row with similar letter(s) do not differ significantly at (P≤0.05) using ls mean Probability difference (PD); Mean separations were based on transformed values of incidence and severity using Arcsine transformation (x+1), but values presented are the untransformed values
² Late blight rating scale of 0-100% (Henfling, 1987);
³ Pooled means across seasons and locations

TABLE 4. Late blight severities and tuber yield of different potato varieties in Kabale and Mbarara

Varieties	No of fields	Late blight severity				Yield (t ha^-1)
		1st visit	2nd visit	3rd visit	Mean	Yield (t ha^-1)
Cruza	1	10.0	30.0	40.0	26.6 b	12.8 bc
Musitamya	1	0.0	5.0	10.0	5.0 d	8.2 d
Rutuku	1	10.0	15.0	20.0	15.0 c	15.1 ab
Sangema	1	25.0	50.0	40.0	38.3 a	16.9 a
Victoria	8	35.0	50.0	60.0	48.3 a	9.7 cd
Mean		16.0	30.0	34.0	26.6	12.5

¹1st visit=vegatative growth; 2nd visit = flowering stage; 3rd visit = Physiological maturity
²Means with the same letters are not significantly different (P<0.05 ) using ls mean probability difference