Epidemiology of parasitic gastrointestinal nematode infections of ruminants on smallholder farms in central Kenya

Research in Veterinary Science 2001, 70, 33–39 doi:10.1053/rvsc.2000.0438, available online at http://www.idealibrary.com on

Epidemiology of parasitic gastrointestinal nematode infections of ruminants on smallholder farms in central Kenya J. M. NGINYI*†, J. L. DUNCAN‡, D. J. MELLOR‡, M. J. STEAR‡, S. W. WANYANGU*, R. K. BAIN*, P. M. GATONGI* *National Veterinary Research Centre, Muguga, KARI, PO Box 32, Kikuyu, Kenya, ‡Department of Veterinary Clinical Studies, Glasgow University Veterinary School, Bearsden, Glasgow, G61 1QH, Scotland

SUMMARY In order to establish the infection pattern with gastrointestinal nematodes in ruminants in the central Kenya highlands, a study was carried out in 58 smallholder farms. The study involved monthly faecal examinations from sheep, goats and cattle and pasture sampling from eight communal grazing areas. Each month, six Dorper worm-free tracer lambs were introduced and four locally grazed cross-bred sheep were purchased for parasite recovery. The mean faecal egg counts (FEC) for cattle were low throughout the study period, whereas those for sheep and goats showed a seasonal pattern with high levels of infection occurring during the two main rainy seasons, especially in March, April and October. There were significant differences in egg counts over time and among farms. Haemonchus contortus was the most prevalent nematode in the tracer lambs whereas the previously exposed locally grazed sheep had significantly lower numbers of H contortus but significantly higher numbers of Trichostrongylus species The highest levels of infection in the tracer lambs occurred in November 1995 and January, May and June 1996. Based on this study, it is now possible to explore the possibility of using strategic treatments for the control of parasitic gastroenteritis in this area of Kenya. ©2001 Harcourt Publishers Ltd

HELMINTH infections in domestic ruminants are of major importance in many agro-ecological zones in Africa and are a primary factor in the reduction of productivity through losses due to mortality and reduced weight gains (Allonby and Urquhart 1975, Fabiyi 1987, Wanyangu and Bain 1994, Gatongi et al 1997). In the agriculturally highpotential areas of Kenya, where ruminant livestock are kept on pasture throughout the year (Maingi et al 1997) and climatic conditions are favourable for the development and survival of free living stages (Dinnik and Dinnik 1958, 1961), infection with parasitic nematodes is important as a limiting factor to production (Wanyangu and Bain 1994). This is further complicated on smallholder farms due to the limited availability of land and other resources. For example, the communal grazing and watering places shared by smallholder farmers are a major source of infection. This is due to the problem of over-stocking associated with these areas and the fact that helminth control is not co-ordinated, the farmers de-worming their animals at different times and with different anthelmintic preparations (Waruiru et al 1993, Mbaria et al 1995, Wanyangu et al 1996, Maingi et al 1997). The resultant increased infection pressure by gastrointestinal helminths is more serious in small ruminants that suffer more from acute disease, particularly haemonchosis. Chronic helminthosis is more widespread and probably of more significance in all grazing ruminants (Allonby and Urquhart 1975) because of its insidious effects which reduce weight gains, milk yield, wool production and carcass quality, especially in situations where nutrition is poor (Gatongi et al 1997).

*Corresponding author. Fax: 00 254 154 32450. 0034-5288/01/010033 + 07 $35.00/0

Infection with gastrointestinal helminths has been extensively studied in many developed countries, especially in Europe, America and Australia, and appropriate times for intervention with anthelmintic drugs have been determined for specific regions. This is not so for many tropical areas, especially those in Africa. Treatment regimes are frequently based on extrapolation of studies carried out elsewhere, and these are often inappropriate due to differences in ecological factors and management practises that exist between different areas. Appropriate information about a given agroecological area and local farming practices are critical issues for consideration in the development of an effective parasite control regime (Gatongi et al 1997). The main aim of this study was to establish the infection pattern of gastrointestinal nematodes in ruminants in smallholder farms in a high potential area in the highlands of central Kenya, and to use this information to design appropriate strategic intervention measures for this and other parts of the country with similar ecological and husbandry characteristics.

MATERIALS AND METHODS Study location The area of study was located in the Mathira division of Nyeri district in central Kenya at an altitude of approximately 2000 metres. This area experiences a bimodal rainfall of between 750 mm and 1500 mm annually. The long rains occur between March and May and the short rains between October and December. The mean monthly minimum temperatures vary from 10˚C to 15˚C and the mean © 2001 Harcourt Publishers Ltd

34

J. M. Nginyi, J. L. Duncan, D. J. Mellor, M. J. Stear, S. W. Wanyangu, R. K. Bair, P. M. Gatongi

monthly maximum temperatures from 20˚C to 25˚C (Sagana State Lodge meteorological station). The topography is characterised by hills and valleys, with streams and rivers at the bottom of the valleys. Dams and empty marshy spaces, which serve as common watering and grazing grounds for livestock, are common in this area. Due to a high population density most of the area has been cleared of natural vegetation to accommodate farm practises. The majority of farm holdings are small with an average of less than five hectares of land. A total of 58 smallholder farms participated in this study, with a collective number of 185 cattle, 110 sheep and 27 goats on these farms. Selection of farms The farms selected for this study were situated along a road transect. Consideration was given to the willingness of the farmer to participate as well as the proximity to the main road. All the farms along the road transect were listed together with the number of sheep, goats and cattle kept on the farms. A total of 100 farms were identified and this formed the sampling frame from which the trial farms were to be selected. The farms were then selected using a set of random numbers generated by a calculator to give a total of approximately 200 cattle and 100 small ruminants, the number considered manageable for the 2-day sampling period allocated for this exercise every month. This number of animals was obtained from a total of 60 smallholder farms and the sheep:goat ratio was 4:1. These farms were visited a second time before final selection to explain to the farmers the objectives of the study and to ascertain whether they would be able to extend the necessary co-operation. Tracer lambs The Dorper lambs used in this study as tracers were bought from Kamogi ranch in Laikipia district when they were approximately three months old and taken to the National Veterinary Research Centre (NVRC), Muguga. They were quarantined under worm free conditions for 1 month during which time they were treated with a levamisolebased anthelmintic (Wormicid®, Cosmos Ltd., Nairobi, Kenya). Six of these lambs were introduced to the communal grazing areas, including roadsides, every month from May 1995 to October 1996. They grazed every day from 09·00 to 16·00 and at night they were housed in a pen with a slatted floor. They had access to tap water ad libitum during housing at night and during the day they watered at the same dams and streams as the local flocks. Permanently grazed sheep Adult ewes bought from farmers in the study area were mainly Red Maasai crosses. The Red Maasai breed is indigenous to East Africa and such cross-bred animals are present on many smallholder farms in the study area and in many other parts of Kenya. The adult ewes were purchased directly from local farmers because those available at the local livestock market had usually come from outside the district. Four ewes were purchased every month.

While indoors at NVRC the sheep were fed on a mixture of grass and lucerne hay, stock pellets (Young stock pencils, Belfast Millers Ltd., Nairobi, Kenya) given as a supplement once a day and water was provided ad libitum. Study design The monthly monitoring for nematode infections of all sheep, goats and cattle in the smallholder farms was carried out between March 1995 and October 1996. For the purpose of analyses, lambs and kids were defined as animals less than 6 months of age and calves were defined as animals less than 12 months of age. After the examination of faecal samples for nematode eggs, those samples that were positive were pooled for each farm and cultured. A differential count was carried out on the larvae harvested from these samples. In addition, herbage samples were collected from eight sites in the study area, larvae harvested, identified and expressed as numbers of infective larvae per kilogram of dry herbage (L3 kg-1 DH). The introductions of tracer lambs and purchase of permanently grazed sheep continued for 18 months, from May 1995 to October 1996. Each month the tracer lambs, which had grazed for one month, and the permanently grazed sheep, were transported to NVRC and housed. Half of each group of sheep, three tracer lambs and two permanently grazed sheep, were slaughtered and necropsied at 3 and 6 weeks, respectively, and differential worm counts carried out. The reason for the extended housing, beyond the prepatent period of most gastrointestinal nematodes, was to enable easier recovery of liver flukes when they were at least 6-weeks old. The results from the fluke component of the study will be reported separately. The parasitological data were examined in conjunction with climatological data from Sagana State Lodge meteorological station, which is located within the study area. Parasitological methods Faecal samples were examined for nematode eggs using the modified McMaster technique (Whitlock 1948), whilst pasture samples were processed and infective larvae differentiated using established methods (MAFF 1986). For both the tracer lambs and permanently grazed sheep, the abomasum was opened along the greater curvature and washed with water under moderate pressure. The washings were made up to 2 litres and a 200 ml sample was taken. Ten aliquots of 4 ml each were examined and the number of worms found multiplied by 50 to give an estimate of the total worm count. Mucosal scrapings from the abomasum were digested with HCl-pepsin at 37˚C; the digests were then made up to 2 litres, sub-samples and aliquots were taken and examined as for the abomasum contents to estimate the size of the mucosal larval population. All samples were preserved by adding 2–3 ml of iodine prior to examination. Similarly, the small intestines of tracer lambs and permanently grazed sheep were opened with gut scissors and the contents washed into a bucket and made up to 2 litres. Samples were taken and examined as above. The contents and mucosa of the large intestine were examined grossly and any parasites present were collected for identification and

RESULTS Faecal egg counts

100 50

Sep-96

Jul-96

May-96

Mar-96

Jan-96

Nov-95

Sep-95

0

Month FIG 2: The rainfall pattern in the study area over the study period and the 10year (1986–1995) mean monthly rainfall figures. Rainfall data were collected from Sagana State Lodge weather station which is within the study area. ■, rainfall; ■ 10-year mean rainfall.

Goats Results of the goat mean FEC are shown in Figure 3. On average, 30 goats were sampled every month. The number of kids available for sampling was so low that the data presented are from adults only. The mean FEC ranged from 138 EPG in August 1995 to 1489 EPG in October 1996. The mean FEC were above 1000 EPG on seven occasions over the 20month study period. The infection pattern in the adult goats tended to be similar but less defined than that observed in the adult sheep but the goat mean FEC were generally higher. Statistical analyses examined the effects of age, animal within farm, farm and time (month of sampling) on the FEC. There were significant differences in FEC (P < 0·0001) between farms and at different times of the year. There were no significant variations due to age, though there were very few kids to include in the analysis or to animals within farms. Cattle The mean FEC for calves and adult cattle in the smallholder farms are shown in Figure 4. The mean FEC for calves were higher than those for the adults though both generally remained low. The FEC were highest in March 1995 and in February and August 1996 in the calves. In the adult cattle the mean FEC were less than 200 EPG in most months and less than 100 EPG in eight of the 20 months of the study period. Statistical analyses examined the effects of age, animal within farm, farm and time (month of sampling) on the FEC. There were highly significant differences in FEC at different times of the year (P<0·0001) with calves having

1500

Mean FEC Month l

b

1000 500

Sep-96

Jun-96

Mar-96

Dec-95

Sep-95

Mar-95

0

Jun-95

3500 3000 2500 2000 1500 1000 500 0

Mar-95 Apr-95 May-95 Jun-95 Jul-95 Aug-95 Sep-95 Oct-95 Nov-95 Dec-95 Jan-96 Feb-96 Mar-96 Apr-96 May-96 Jun-96 Jul-96 Aug-96 Sep-96 Oct-96

Mean FEC

Sheep The results of faecal examination for nematode eggs in lambs and adult sheep in the smallholder farms are shown in Figure 1 and the rainfall pattern over the same period is shown in Figure 2. The mean faecal egg counts for lambs and adults showed a similar pattern over the study period but those from the lambs were generally higher than those from the adults. On a few occasions, i.e. August 1995, January and March 1996, the lambs and adult sheep had almost similar mean faecal egg count (FEC). For the majority of the months the mean FEC for lambs were below 2000 EPG. The times when the mean FEC exceeded 2000 EPG were April 1995 and February and October 1996. The adult sheep had mean FEC generally below 1000 EPG in all months except February and October 1996. The lowest mean FEC of less than 300 EPG for both groups of sheep were recorded in August 1995 and January 1996. Statistical analyses examined the effects of age, animal within farm, farm and time (month of sampling) on the FEC. There were highly significant differences in FEC between farms and at different times of the year (P < 0·0001) with lambs having significantly higher FEC than adult sheep (P < 0·05). There was no significant variation attributable to individual animals within farms.

150

Jul-95

Mixed model repeated measures analysis of variance to test the effects of various factors were carried out with the Mixed and General Linear Models (GLM) programmes on the SAS package (Statistical Analysis Systems Institute, Cary, North Carolina, USA). Faecal egg counts were logarithm-transformed in the form of log10(EPG +1) to stabilise the variance before analysis. Total worm counts as well as the individual parasite species were compared between the tracer lambs and the permanently grazed sheep. The effects of time and duration of housing on worm burdens were also tested.

200

May-95

Statistical analyses

35

250

Mar-95

counting. The parasites from the large intestine were not included in the statistical analyses.

Rainfall (mm)

Parasitic gastrointestinal infections on farms in Kenya

Month

d l

FIG 1: Mean monthly faecal egg counts (FEC) for lambs and adult sheep in smallholder farms. ◆ , FEC lambs; ■ FEC adults.

FIG 3: Mean monthly faecal egg counts (FEC) for adult goats in the smallholder farms.

J. M. Nginyi, J. L. Duncan, D. J. Mellor, M. J. Stear, S. W. Wanyangu, R. K. Bair, P. M. Gatongi

36

(33 per cent), Trichostrongylus (32 per cent), Oesophagostomum (20 per cent), Cooperia (13 per cent) and Strongyloides (2 per cent).

Mean FEC

800 600 400 200

Total worm counts Sep-96

Jul-96

May-96

Mar-96

Jan-96

Nov-95

Sep-95

Jul-95

May-95

Mar-95

0

Month FIG 4: Mean monthly faecal egg counts (FEC) for calves and adult cattle in smallholder farms. ◆ , FEC calves; ■ FEC adults.

significantly higher FEC than adult cattle (P < 0·05). Variation in FEC in both calves and adult cattle due to farm were also significant (P < 0·05). Pasture larval counts The results of mean larval recoveries from herbage taken from the eight sampling sites are shown in Figure 5. The peaks of pasture infectivity occurred in July, August and December 1995, and in February and March 1996. Overall, the majority of the larvae recovered from herbage were those of Haemonchus spp. (71·2 per cent) followed by Trichostrongylus spp. (14·6 per cent), Cooperia spp. (10·4 per cent) and Oesophagostomum spp (3·8 per cent). Statistical analyses showed that there were highly significant effects of time (month of sampling) on the total number of infective larvae recorded (P < 0·0001) but there were no significant differences between the different sites where herbage samples were taken. Differential larval counts from coprocultures Infective larvae harvested from pooled coprocultures from the sheep samples were from the following genera of nematodes: Haemonchus (33 per cent), Trichostrongylus (29 per cent), Oesophagostomum (25 per cent), Cooperia (7 per cent), and Strongyloides (6 per cent). In the limited coprocultures from goats, the following genera of nematodes were present: Haemonchus (24·9 per cent), Trichostrongylus (50·0 per cent), Oesophagostomum (14·0 per cent), Cooperia (8·4 per cent) and Strongyloides (2·7 per cent). The distribution of nematode larvae harvested from cattle faeces according to genera were as follows: Haemonchus

L3 kg–1 DH

1000 800 600 400 200

Sep-96

Jun-96

May-96

Mar-96

Jan-96

Nov-95

Sep-95

Jul-95

May-95

Mar-95

0

Month FIG 5: area.

Mean larval counts from herbage samples collected from the study

Tracer lambs Table 1 shows the means of total worm counts from the tracer lambs over the 18-month period. Haemonchus contortus was the most predominant nematode recovered from the tracers. This was followed by Trichostrongylus axei, T colubriformis and Cooperia spp. In the majority of the months, Haemonchus was the more prevalent. Counts for Oesophagostomum columbianum were not included in these results because these parasites may have been present in the lambs when they were bought and the anthelmintic used to treat the lambs during quarantine was found to be ineffective in eliminating the larvae of Oe columbianum in lesions of pimply gut. The total counts of these parasites were very low and never exceeded 50 per sheep. The tracer lambs acquired low burdens of nematode parasites from pasture in May, June and July 1995 and in August, September and October in 1996. These periods coincided with the periods of low rainfall (Fig 2). The months with the highest mean worm counts (more than 2000) were November 1995 and January, May and June 1996. Overall the distribution of the different nematode species recovered from the tracer lambs were as follows: H contortus (55·6 per cent), T axei (16·7 per cent), T colubriformis (20·7 per cent), Cooperia spp. (3·6 per cent), Oesophagostomum spp. (3·1 per cent) and Trichuris spp. (0·3 per cent). Statistical analyses examined the effects of time (month of grazing) and duration of housing after grazing (3 or 6 weeks) on the total worm burdens and differences in worm species at necropsy. There were highly significant differences due to time (P < 0·0001) on the total worm burdens. The duration of housing did not influence the total worm burdens (P > 0·05). Local permanently grazed sheep The results of the mean total worm counts for the local permanently grazed sheep each month are shown in Table 2. The permanently grazed sheep harboured nematode infections throughout the study period, including the months that had very low rainfall (Fig 2). The months with the highest peaks of infection (>2000 worms) were May, September and November 1995 and January, February, May and August 1996. The lowest mean recoveries (<1000 worms) were made in June, July and December 1995 and in September 1996. In the majority of the months Trichostrongylus spp. were more abundant than H contortus. The overall distribution of different nematode species recovered from the permanently grazed sheep were as follows: H contortus (19·6 per cent), T axei (36·7 per cent), T colubriformis (36·8 per cent), Cooperia spp. (5·0 per cent), Oesophagostomum spp. (1·1 per cent) and Trichuris spp. (0·3 per cent). Statistical analyses showed a highly significant effect of time of purchase of the permanently grazed sheep on worm burdens (P < 0·0001) but the duration of housing (3 or 6 weeks) had no effect. Statistical comparisons between the two types of sheep showed that the tracer lambs had significantly higher H contortus burdens than the permanently

Parasitic gastrointestinal infections on farms in Kenya

37

TABLE 1: Mean worm counts from the six Dorper tracer lambs introduced into the communal grazing areas every month Time

Mean worm counts from tracer lambs H contortus

May 1995 June July Aug Sept Oct Nov Dec Jan 1996 Feb Mar Apr May June July Aug Sept Oct

542 550 420 717 375 342 2990 683 2050 560 1183 975 2183 1350 1267 210 100 33

T axei 25 0 130 500 375 317 1130 183 400 150 292 125 208 417 642 50 75 0

T colubriformis 150 0 70 708 442 300 910 667 1158 100 325 133 233 867 100 0 0 0

Cooperia spp 10 0 0 17 25 8 600 0 125 0 0 8 58 125 83 10 0 0

Total 727 550 620 1942 1217 967 5630 1533 3733 810 1800 1241 2682 2759 2092 270 175 33

TABLE 2: Mean worm counts recovered from local permanently grazed sheep purchased from farms in the study area Time

Mean worm counts from permanently grazed sheep H contortus

May 1995 June July Aug Sept Oct Nov Dec Jan 1996 Feb Mar Apr May June July Aug Sept Oct

100 650 325 313 25 150 1150 63 75 388 33 517 2238 625 413 913 117 117

T axei 1350 0 488 288 1200 867 567 750 2513 3138 233 523 1088 213 650 1150 0 317

grazed sheep (P < 0·0001) and the reverse was observed for T axei. The differences in total worm burdens and in T colubriformis and Cooperia spp. burdens between the tracer lambs and permanently grazed sheep were not significant (P > 0·05).

DISCUSSION The results of faecal examinations from sheep, goats and cattle in the smallholder farms in this study showed that infection with gastrointestinal nematodes were present throughout the year. However, the levels of infection were different for small ruminants and cattle. The mean FEC for calves, though higher than those from the adult cattle, remained low throughout the study. On the basis of FEC, the infection levels in adult cattle were far below the minimum of 500 EPG commonly recommended for anthelmintic treatment in tropical areas (Hansen and Perry 1994). In the sheep (lambs and adults) the mean FEC were above 500 EPG for

T colubriformis 800 0 100 963 2588 250 1900 0 3813 1450 583 367 275 163 425 388 0 767

Cooperia spp. 50 0 0 25 25 33 50 0 563 700 417 0 50 13 13 0 0 0

Total 2304 650 913 1509 3838 1300 3667 813 6964 5676 1266 1407 3651 1014 1501 2451 117 1201

most months, with distinct peaks of infection occurring during the wetter times of the year, especially during the months of April and October 1996. These results were in agreement with other studies which showed that rainfall was an important factor in determining levels of infection (Southcott et al 1976, Waruiru et al 1993, Gatongi et al 1997, Wanyangu et al 1997). The results from the goats FEC, though fewer in number, showed a less defined seasonal pattern compared with the sheep. In this study the problem of infection with gastrointestinal nematodes was found to be of greater importance in small ruminants than in cattle. The findings from faecal examination of the cattle, sheep and goats in the smallholder farms were reflected in the recovery of infective larvae from pasture samples. High levels of pasture contamination with L3 were associated with periods of high rainfall, and this was especially evident during the two main rainy seasons. Low numbers of larvae were generally recovered from pasture during the intervening dry periods. The results also showed that Haemonchus was the most abundant nematode genus on pasture

38

J. M. Nginyi, J. L. Duncan, D. J. Mellor, M. J. Stear, S. W. Wanyangu, R. K. Bair, P. M. Gatongi

throughout the study period. This is probably related to the high fecundity of Haemonchus which means that it is likely to be recovered from pasture in higher numbers than larvae of other genera (Gibson and Everett 1976). However, this study showed that Trichostrongylus spp. might play a more important role in the aetiology of parasitic gastroenteritis syndrome of sheep and goats than was previously thought. This view was supported by findings from the necropsy worm counts from the locally grazed sheep bought from the study area. Both Trichostrongylus species, T axei and T colubriformis, were recovered in significantly higher proportions than H contortus in the majority of months. The reasons for this were not immediately clear, especially when both herbage larval counts and tracer worm burdens had indicated that Haemonchus was likely to be much more abundant than Trichostrongylus. There are several possible reasons for this finding. One possibility is that the locally grazed sheep, which were Red Maasai crosses, were capable of suppressing the establishment of H contortus infection, as this has been shown previously in pure Red Maasai sheep (Preston and Allonby 1978, 1979, Mugambi et al 1997). The fact that H contortus was more abundant than both T axei and T colubriformis in the tracer lambs used throughout the study period is probably related in part to the fact that these were Dorpers, which is a breed that has been shown to be highly susceptible to H contortus infection (Preston and Allonby 1978, 1979, Mugambi et al 1997). Furthermore, the tracer lambs were younger than the permanently grazed sheep and the development of immunity to trichostrongylid infections has been shown to be related to age among other factors (Gamble and Zajac 1992). Another possibility for the above findings is that in sheep with an acquired immunity to Haemonchus, there may be less competition for other nematode species to colonise the gastrointestinal tract. This may have some implications in potential breeding programmes for resistance to H contortus in that resistant sheep may suffer heavier infections of other nematodes. The different aspects of helminth infections investigated in this study gave an insight into their seasonality in the central Kenya highlands. The results from FEC of both sheep and cattle, as well as those from the pasture larval counts, showed the trend of infection with parasitic gastrointestinal nematodes in this area. The use of tracer lambs and permanently grazing animals has been considered to be a better method than pasture sampling for monitoring the availability of infective larvae on pasture and therefore the seasonality of infection with parasitic nematodes (Uriarte and Valderrabano 1989, Wanyangu et al 1997). However, in this study the permanently grazed animals harboured infections throughout, although the levels of infection varied significantly with month of sampling. The pattern of infection was not as distinct as in the susceptible young tracer lambs, although it was clear that heavier infections in the permanently grazed sheep were generally observed during the rainy seasons, with the exception of a few months where relatively high mean worm counts were recorded during periods of low rainfall. This was perhaps not surprising considering that these sheep were probably carrying infections picked up over prolonged periods prior to their purchase. For this study however, one clear observation was that sheep in the study area had substantial nematode worm

burdens at most times of the year. In contrast, the worm burdens of the tracer lambs reflected infections acquired from pasture during the single month that they had been grazing and the results from these animals were therefore more likely to indicate the availability of infective larvae to grazing animals in the study area at different times of the year. The levels of infection in the tracer lambs appeared to be closely related to the rainfall pattern, the highest levels occurring as peaks in January and November. These findings suggested that strategic anthelmintic control of PGE in sheep would be feasible in the highlands of central Kenya. It is recommended that onfarm trials based on these results be carried out to compare a strategic regime based on these results with the existing control measures in central Kenya.

ACKNOWLEDGEMENTS This study was supported by funds from the Livestock Helminthology Project under phase two of the National Agricultural Research Project (NARP II), a joint programme by the Department for International Development (DFID) of the British government and the Kenya government through KARI. We acknowledge with thanks, all those farmers in Nyeri district who allowed us to use their animals and other facilities for this study. We also thank the members of staff from Karatina VIL, veterinary extension staff in Mathira division, the DVO, Nyeri and all the technical staff at the helminthology laboratory at the NVRC for their valuable input to this work.

REFERENCES ALLONBY, E. W. & URQUHART, G. M. (1975) The epidemiology and pathogenic significance of haemonchosis in a Merino flock in East Africa. Veterinary Parasitology 1, 129–143 DINNIK, J. A. & DINNIK, N. N. (1958) Observations on the development of Haemonchus contortus larvae under field conditions in the Kenya highlands. Bulletin of Epizootic Diseases in Africa 6, 11–21 DINNIK, J. A. & DINNIK, N. N. (1961) Observations on the longevity of Haemonchus contortus larvae on pasture herbage in the Kenya Highlands. Bulletin of Epizootic Diseases in Africa 9, 193–208 FABIYI, P. J. (1987) Production losses and control of helminths in ruminants of tropical region. International Journal for Parasitology 17, 435–540 GAMBLE, H. R. & ZAJAC, A. M. (1992) Resistance of St. Croix lambs to Haemonchus contortus in experimentally and naturally acquired infections. Veterinary Parasitology 41, 211–225 GATONGI, P. M., SCOTT, M. E., RANJAN, S., GATHUMA, J. M., MUNYUA, W. K., CHERUIYOT, H. & PRICHARD, R. K. (1997) Effects of three nematode anthelmintic treatment regimes on flock performance of sheep and goats under extensive management in semi-arid Kenya. Veterinary Parasitology 68, 323–336 GIBSON, T. E. & EVERETT, G. (1976) The ecology of the free-living stages of Haemonchus contortus. British Veterinary Journal 132, 50–59 HANSEN, J. & PERRY, B. (1994) The Epidemiology, Diagnosis and Control of Helminth Parasites of Ruminants 2nd ed. Nairobi: International Laboratory for Research on Animal Disease, pp 171 MAFF. (1986) Manual of Veterinary Parasitological Techniques. Technical Bulletin No. 18. London: HMSO, pp 129 MAINGI, N., GICHANGA, E. J. & GICHOHI, V. M. (1993) Prevalence of gastrointestinal helminths and coccidia parasites and frequency distribution of some nematode genera of goats on some farms in four districts of Kenya. Bulletin of Animal Health and Production in Africa 41, 285–290 MAINGI, N., BJORN, H., THAMSBORG, S. M., MUNYUA, W. K., GATHUMA, J. M. & DANGOLLA, A. (1997) Worm control practices on sheep farms in Nyandarua District of Kenya. Acta Tropica 68, 1–9 MBARIA, J. M., MCDERMONT, J. M., KYULE, M. N., GICHANGA, E. J. & MUNYOLE, P. N. (1995) Prevalence of helminths and helminth control practices in Molo division, Nakuru district, Kenya. Bulletin of Animal Health and Production Africa 43, 35–44 MUGAMBI, J. M., BAIN, R. K., WANYANGU, S. W., IHIGA, M. A., DUNCAN, J. L., MURRAY, M. & STEAR, M. J. (1997) Resistance of four sheep breeds to

Parasitic gastrointestinal infections on farms in Kenya natural and subsequent artificial Haemonchus contortus infections. Veterinary Parasitology 69, 265–273 PRESTON, J. M. & ALLONBY, E. W. (1978) The influence of breed on the susceptibility of sheep and goats to a single experimental infection with Haemonchus contortus. Veterinary Record 103, 509–512 PRESTON, J. M. & ALLONBY, E. W. (1979) The influence of breed on the susceptibility of sheep to Haemonchus contortus in Kenya. Research in Veterinary Science 26, 134–139 SOUTHCOTT, W. H., MAJOR, G. W. & BARGER, I. A. (1976) Seasonal pasture contamination and availability of nematodes for grazing sheep. Australian Journal of Agricultural Research 27, 277–286 URIATE, J. & VALDERRABANO, J. (1989) An epidemiological study of parasitic gastroenteritis in sheep under an intensive grazing system. Veterinary Parasitology 31, 71–81 WANYANGU, S. W. & BAIN, R. K. (1994) The impact of helminth infection on small ruminant production in Tropical Africa. The Kenya Veterinarian 18 102–106

39

WANYANGU, S. W., BAIN, R. K., RUGUTT, M. K., NGINYI, J. M. & MUGAMBI, J. M. (1996) Anthelmintic resistance amongst sheep and goats in Kenya. Preventive Veterinary Medicine 25, 285–290 WANYANGU, S. W., KARIMI, S., MUGAMBI, J. M. & BAIN, R. K. (1997) Availability of Haemonchus controtus L3 larvae on pasture at Kiboko: a semi-arid warm agro-climatic zone in Kenya. Acta Tropica 68, 183–189 WARUIRU, R. M., MBUTHIA, P. G. & KIMORO, C. O. (1993) Prevalence of gastrointestinal parasites and liver flukes in calves in Mathira Division of Nyeri District, Kenya. Bulletin of Animal Health and Production in Africa 41, 291–296 WHITLOCK, H. V. (1948) Some modification of the McMaster egg counting technique and apparatus. Journal of the Council of Scientific and Industrial Research in Australia 21, 177–180

Accepted November 15, 2000