- Email: [email protected]
Response to helminth infection of sheep selected for resistance to Haemonchus contortus
0
RESPONSE
002~7519/90 $3.043+ 0.00 Pergamon Press p/c for Porasiro~ogy
Society
I990 Ausrrolion
TO HELMINTH INFECTION OF SHEEP SELECTED RESISTANCE TO HAEMONCHUS CONTORTUS R. R. WOOLASTON,* CSIRO
Pastoral
Research
Laboratory,
I.
A. BARGER Armidale,
FOR
and L. R. PIPER
New South Wales 2350, Australia
(Received 19 March 1990; accepted 16 June 1990) R. R., BAKER I.A. and PIPER L. R. 1990. Response to helminth infection of sheep selected for resistance to Haemonchus contortus. International Journalfor Parasitology 20: 1015-1018. Lines of Merino sheep selected for increased (IRH) and decreased (DRH) resistance to Huemonchus contortus were compared with an unselected (CH) line, after approximately four generations of selection. Measurements were recorded on 69 IRH, 47 DRH and 84 CH animals. Following artificial challenge with H. contortus, the IRH line had significantly (PC 0.001) lower faecal egg counts than the CH and DRH lines (2730, 12,720 and 17,400 epg, respectively). Significant differences (PC 0.05) were found between all lines in the minimum packed cell volumes during artificial infection (25.7,22.0 and 20.3%) and in faecal egg counts after natural infection (140, 3590 and 8750 epg). Differences were also recorded (PcO.05) following artificial challenge with Trichostrongylus colubriformis (490,840 and 1340 epg). On a percentage basis, faecal egg counts in the IRH line deviated less from the CH line following artificial infection with T. colubriformis (42%) than with H. contorfus (79%). The reverse was true for the DRH line (60 and 37%, respectively). Differences in egg AbShCt-WOOLASTON
output of this magnitude should have a marked effect on requirements for anthelmintic treatment, rate of development of drug resistance and level of pasture contamination when the lines are grazed separately. INDEX KEY WORDS: Genetic Trichostrongylus colubriformis.
resistance: sheep; nematodes;
INTRODUCTION to H. contortus
has been demonstrated in Merino sheep, with estimates of heritability found to be around 0.3 (Piper, 1987; Albers, Gray, Piper, Barker, Le Jambre & Barger, 1987). Based on earlier preliminary estimates (Le Jambre, 1978) a selection experiment was established comprising lines of Merinos selected for increased and decreased resistance together with a random-bred control line. This paper reports the differences between the lines that have emerged in response to artificial and natural challenge with H. contortus larvae and the effect of this selection on resistance to T. colubriformis. GENETIC variation
in resistance
MATERIALS
AND METHODS
Selection procedure. Three lines of sheep were established in 1977, each with 100 mixed age (2-6 years) fine-wool Merino ewes and their 1976-born progeny. All foundation animals were drawn from a larger flock of 2000 ewes. Ewes born from 1972 to 1975 were allocated to the three lines at random. Animals born in 1976 and 1977 were allocated to the increased resistance (IRH) and decreased resistance (DRH) lines on the basis of their packed cell volume (PCV) decline following artificial challenge with H. contortus larvae, while those in the random-bred (CH) line were allocated at random. From 1978 onwards, replacement animals were
* To whom all correspondence
should be addressed.
immunity; Haemonchus
contortus;
chosen within the DRH and IRH lines according to their maximum faecal egg counts following artificial challenge with H. contortus larvae. Replacement ewes and rams in the CH line were chosen at random. Each line was maintained
with five ewe age groups and one ram age group. Thus approximately 23 new l&month-old ewes and five new rams (18 months old, one from each sire group born) entered their respective breeding flock each year. These were drawn from approximately 38 candidates of each sex per line. Lambs born in 1988 represented the equivalent of just under four generations of selection. Management. The three lines grazed together as one flock except for 5 weeks at joining and a further 5 weeks at lambing, when they were run in single-sire mating groups. Weaners from all lines were run together as one flock until the sexes were separated at about 8 months of age. Except during the challenge period, effective anthelmintics were used for parasite control in a regime typically used by commercial sheep producers in the area. Challenge. The experimental procedure was similar each year. During early spring, lambs were born into one of 15 small paddocks, according to their sire group. Weaning occurred some 3 months after the completion of lambing, when lambs were drenched with a broad-spectrum anthelmintic and moved to uncontaminated pasture. At 5-6 months of age, weaners were orally infected with 10,000 H. contortus larvae and faecal samples taken 3, 4 and 5 weeks later. Faecal counts were determined using a modified McMaster method with a lower limit ofdetection of 100 epg. Blood samples were also taken for PCV measurements at the beginning of infection and at the times of faecal sampling. In this report, the maximum epg and minimum PCV during the
1015
R. R. WOOLASTON, I. A BARGER and L. R. PIPER
1016
post-infection period were the traits used. Additional observations were made on the 1988-born weaners. After the artificial challenge period all animals were drenched with ivermectin at the manufacturer’s recommended dose rate, then returned to graze contaminated pasture. Ram and ewe weaners were held in similar adjoining paddocks. Following 6 weeks of natural challenge, epg and PCVs were recorded, then weaners were drenched with ivermectin and closantel at the recommended dose rates. Larval differentiation of faecal samples bulked within each selection line revealed that the larvae acquired during the natural infection period were almost exclusively H. contorfus. In order to test the resistance of each line to Trichosrrongylus species, all weaners were given an oral infection of 10,000 T. colubriformis larvae I week after ivermectin and closantel treatment. Faecal egg counts were again recorded 4 weeks after this infection. Various strains of H. contortus larvae have been used over the years of the selection experiment. For the data reported here, the strain used for artificial infection originated from a field isolate obtained at the University of New England’s Kirby Research Farm, Armidale. The T. colubriformis strain was provided by the CSIRO McMaster Laboratory, Sydney. Statistical analysis. For these analyses, data were collected from 69 IRH, 47 DRH and 84 CH line weaners. Because of the unbalanced nature of the data, the method of least squares was used for the analysis of variance of epg and PCVs. The least squares means presented are thus free of any bias caused by the unbalanced design. Faecal egg counts were transformed prior to analysis to remove skewness and kurtosis. Effects tested in the analysis of variance were selection line, sex, birth status (twin or single), age of dam, age of infection and first order interactions. With the
exception of line effects, all main effects and interactions found to be not significant were omitted from the final model. A second model was fitted to test the significance of the line x challenge type interaction. For this analysis, the faecal egg counts following each challenge were treated as repeat observations on each animal. The effects of line, animal within line, challenge type and the line x challenge type interaction were then tested using least-squares analysis of variance. In order to remove scale effects when testing the interaction, log-transformed egg counts were expressed in
standard deviation units. RESULTS Line effects were highly significant (P < 0.001) for all traits except PCV of uninfected animals (Table 1). Sex effects were only significant for PCV following natural infection (P
35% 30% 25%
Packed
20%
cell volume
15% 10% 5% 0% <
Uninfected
Artificial challenge
Natural challenge
Selection line
I
IRH
CH
DRH
FIG. 1. Mean packed cell volumes (PCV) in the increased resistance (IRH), decreased resistance (DRH) and control (CH) lines prior to infection, during artificial infection with Haemonchus larvae and after a period of natural infection. All differences were significant (PcO.05) except prior to infection and between the CH and DRH lines after natural infection.
Genetic
of sheep to H. contortus
resistance
TABLE~--SUMMARY OFANALYSES OFVARIANCE FORPACKED CELL VOLUMES (%) IN UNINFECTED
ANIMALS
AND IN THE SAME ANIMALS
FOLLOWINGARTIFICIALANDNATURALCHALLENGEWITH
H.COntOrtus
Source of variation Line Sex Error Packed cell volumes-mean Uninfected Artificial H. contortus Natural H. contortus Faecal egg Artificial Natural Artificial
squares 14.9749 452.9555** 241.8353**
counts-mean squares H. contortus 11.8574** H. contortus 56.7018** T. colubriformis 2.7675**
Degrees of freedom
11.0667 3.6875 96.0395*
10.2239 13.9983 17.2580
0.0006 0.0973 0.0575
0.2423 1.2045 0.1732
1
196
2
Mean squares for faecal egg counts (in log,,, eggs per gram) following three challenge periods are also tabulated. * P
The patterns in faecal egg counts corresponded with the levels of anaemia (Fig. 2). Maximum epg during artificial H. contortus challenge were 2730 in the IRH line (back-transformed), compared with 12,720 in the CH line and 17,400 in the DRH line. Although the comparison of IRH animals with CH and DRH animals was highly significant (P < O.OOl), the
1017
difference between CH and DRH means was not significant. After natural challenge, the corresponding means were 140, 3590 and 8750 epg, with all three means significantly different from each other (PC 0.05). Following artificial challenge with T. colubriformis larvae, the least squares means were 490,840 and 1340 epg, respectively (Fig. 2), again all significantly different from each other (P~0.05). In all cases except initial PCV, the CH line mean was intermediate to those of the divergent selection lines, but there was an interaction between selection line and parasite species (PC 0.01, not tabulated). Following artificial infection with H. contortus, faecal egg counts in the DRH line deviated further from the CH line (- 79%) than the DRH line (+37%). However, following infection with T. colubriformis, the respective deviations were - 42% and + 60%. DISCUSSION
These results demonstrate that selective breeding of Merinos for resistance to artificial challenge with infective H. contortus larvae is effective. Substantial selection progress has been made within a few generations, as was predicted from the estimates of genetic variation in unselected flocks (reviewed by Piper, 1987). They also parallel the findings of Windon, Dineen & Wagland (1987) who have demonstrated genetic differences in Merinos selected on the basis of
_
Artificial
Natural
H. contortus
Artificial
H. contorhrs
T. cdubtiformi~
Selection line I
IRH
CH
1500
4
1
DRH
FIG. 2. Mean worm egg counts in the increased resistance (IRH), decreased resistance (DRH) and control (CH) lines, following artificial and natural H. contortus (Hc) challenge and artificial T. colubriformis (Tc) challenge. All differences were significant (PcO.05) except between CH and DRH following artificial infection with H. contortus.
1018
R. R. WOOLASTON,I. A. BARGERand L. R. PIPER
artificial challenge following vaccination with irradiated T. colubriformis larvae. Baker, Watson & Harvey (1988) also reported progress when selecting Romney sheep for low epg in an environment where Trichostrongylus and Ostertagia were the most prevalent genera. A most encouraging aspect of this study is that the differences in epg between the lines persisted after grazing naturally infected pastures. Simulation studies predict that differences of this magnitude should have a marked effect on requirements for anthelmintic treatment, rate of development of drug resistance and level of pasture contamination (Barger, 1989). The experimental requirements of the selection programme have precluded weaners from each line being run separately, so we have as yet been unable to verify these predictions in the field. The other notable aspect is that selection for resistance to H. contortus has also conferred a degree of resistance to T. colubriformis. On a percentage basis, faecal egg counts in the IRH line deviated less from the CH line following artificial infection with T. colubriformis (42%) than with H. contortus (79%). Surprisingly, the reverse was true for the DRH line (60 and 37%, respectively). This interaction is somewhat puzzling and could be an artifact of either the method used to assess resistance or the method of transformation. However, the latter explanation is unlikely, because similar results were obtained in preliminary analyses using no transformation or a square root transformation. Evidence for a genetic correlation between resistance to the two parasites was also reported by Windon et al. (1987), who found that moderate levels of protection against H. contortus were detectable in the third generation of selection for resistance to T. colubriformis. In another study, Gray & Barger (cited by Barger, 1989) compared the progeny of normal Merino sires with progeny of a sire known to confer a high level of resistance to H. contortus. They found that the lambs genetically resistant to H. contortus also displayed a degree of resistance to natural infections of Trichostrongylus and Ostertagia species. It is possible that the prior exposure of lambs to a high level of H. contortus in this study may have affected the magnitude of response to T. colubriformis, but there is probably also an effect of ageper se. In the study of Gray & Barger (cited above), differences in
resistance between the genotypes disappeared by 8-9 months of age, indicating that the age at which comparisons are made may be critical. It is not clear from our results whether the lines differ primarily in their ability to develop resistance or whether they merely differ in the age at which resistance develops. Nor is it apparent whether the differences in resistance to T. colubriformis were a consequence of the immunity developed to H. contortus or simply another manifestation of the speed at which the lines can mount an immune response. These questions can only by resolved by serial observations made under carefully controlled conditions. Acknowledgements-The significant contributions of Dr L. Le Jambre during the earlier years of this study are gratefully acknowledged. Thanks are also due to A. Lisle, G. Brown, B. Dennison, R. Nethery and G. Uphill for their skilled and enthusiastic technical assistance. Financial assistance for the maintenance of these flocks has been provided by the Australian Wool Corporation. REFERENCES ALBERSG. A. A., GRAYG. D., PIPERL. R., BARKERJ. S. F., LE JAMBREL. F. & BARCER I. A. 1987. The genetics of resistance and resilience to Haemonchus contortus in young Merino sheep. International Journal for Parasitology 17: 1355-1363. BAKERR. L., WATSONT. G. & HARVEYT. G. 1988. Genetic variation in, and selection for, resistance or tolerance to internal nematode parasites in sheep. Proceedings, 3rd World Congress Sheep and Beef Cattle Breeding, INRA, Paris, Vol. I, pp. 637-639. BARGER I. A. 1989. Genetic resistance of hosts and its influence on epidemiology. Veterinary Parasitology 32: 2135. LE JAMBRE L. F. 1978. Host genetic factors in helminth control. In: The Epidemiology and Control of Gastrointestinal Parasites of Sheep in Australia (Edited by DONALD A. D., SOUTHCO~TW. H. and DINEEN J. K.), pp. 137-141. CSIRO, Melbourne. PIPERL. R. 1987. Genetic variation in resistance to internal parasites. In: Merino Improvement Programs in Australia (Edited by MCGU~RK B. J.), pp. 351-363. Australian Wool Corporation, Melbourne. WINDONR. G., DINEENJ. K. & WAGLANDB. M. 1987. Genetic control of immunological responsiveness against the intestinal nematode Trichostrongylus colubrlformis in lambs. In: Merino Improvement Programs in Australia (Edited by MCGUIRK B. J.), pp. 371-375. Australian Wool Corporation, Melbourne.
RESPONSE
002~7519/90 $3.043+ 0.00 Pergamon Press p/c for Porasiro~ogy
Society
I990 Ausrrolion
TO HELMINTH INFECTION OF SHEEP SELECTED RESISTANCE TO HAEMONCHUS CONTORTUS R. R. WOOLASTON,* CSIRO
Pastoral
Research
Laboratory,
I.
A. BARGER Armidale,
FOR
and L. R. PIPER
New South Wales 2350, Australia
(Received 19 March 1990; accepted 16 June 1990) R. R., BAKER I.A. and PIPER L. R. 1990. Response to helminth infection of sheep selected for resistance to Haemonchus contortus. International Journalfor Parasitology 20: 1015-1018. Lines of Merino sheep selected for increased (IRH) and decreased (DRH) resistance to Huemonchus contortus were compared with an unselected (CH) line, after approximately four generations of selection. Measurements were recorded on 69 IRH, 47 DRH and 84 CH animals. Following artificial challenge with H. contortus, the IRH line had significantly (PC 0.001) lower faecal egg counts than the CH and DRH lines (2730, 12,720 and 17,400 epg, respectively). Significant differences (PC 0.05) were found between all lines in the minimum packed cell volumes during artificial infection (25.7,22.0 and 20.3%) and in faecal egg counts after natural infection (140, 3590 and 8750 epg). Differences were also recorded (PcO.05) following artificial challenge with Trichostrongylus colubriformis (490,840 and 1340 epg). On a percentage basis, faecal egg counts in the IRH line deviated less from the CH line following artificial infection with T. colubriformis (42%) than with H. contorfus (79%). The reverse was true for the DRH line (60 and 37%, respectively). Differences in egg AbShCt-WOOLASTON
output of this magnitude should have a marked effect on requirements for anthelmintic treatment, rate of development of drug resistance and level of pasture contamination when the lines are grazed separately. INDEX KEY WORDS: Genetic Trichostrongylus colubriformis.
resistance: sheep; nematodes;
INTRODUCTION to H. contortus
has been demonstrated in Merino sheep, with estimates of heritability found to be around 0.3 (Piper, 1987; Albers, Gray, Piper, Barker, Le Jambre & Barger, 1987). Based on earlier preliminary estimates (Le Jambre, 1978) a selection experiment was established comprising lines of Merinos selected for increased and decreased resistance together with a random-bred control line. This paper reports the differences between the lines that have emerged in response to artificial and natural challenge with H. contortus larvae and the effect of this selection on resistance to T. colubriformis. GENETIC variation
in resistance
MATERIALS
AND METHODS
Selection procedure. Three lines of sheep were established in 1977, each with 100 mixed age (2-6 years) fine-wool Merino ewes and their 1976-born progeny. All foundation animals were drawn from a larger flock of 2000 ewes. Ewes born from 1972 to 1975 were allocated to the three lines at random. Animals born in 1976 and 1977 were allocated to the increased resistance (IRH) and decreased resistance (DRH) lines on the basis of their packed cell volume (PCV) decline following artificial challenge with H. contortus larvae, while those in the random-bred (CH) line were allocated at random. From 1978 onwards, replacement animals were
* To whom all correspondence
should be addressed.
immunity; Haemonchus
contortus;
chosen within the DRH and IRH lines according to their maximum faecal egg counts following artificial challenge with H. contortus larvae. Replacement ewes and rams in the CH line were chosen at random. Each line was maintained
with five ewe age groups and one ram age group. Thus approximately 23 new l&month-old ewes and five new rams (18 months old, one from each sire group born) entered their respective breeding flock each year. These were drawn from approximately 38 candidates of each sex per line. Lambs born in 1988 represented the equivalent of just under four generations of selection. Management. The three lines grazed together as one flock except for 5 weeks at joining and a further 5 weeks at lambing, when they were run in single-sire mating groups. Weaners from all lines were run together as one flock until the sexes were separated at about 8 months of age. Except during the challenge period, effective anthelmintics were used for parasite control in a regime typically used by commercial sheep producers in the area. Challenge. The experimental procedure was similar each year. During early spring, lambs were born into one of 15 small paddocks, according to their sire group. Weaning occurred some 3 months after the completion of lambing, when lambs were drenched with a broad-spectrum anthelmintic and moved to uncontaminated pasture. At 5-6 months of age, weaners were orally infected with 10,000 H. contortus larvae and faecal samples taken 3, 4 and 5 weeks later. Faecal counts were determined using a modified McMaster method with a lower limit ofdetection of 100 epg. Blood samples were also taken for PCV measurements at the beginning of infection and at the times of faecal sampling. In this report, the maximum epg and minimum PCV during the
1015
R. R. WOOLASTON, I. A BARGER and L. R. PIPER
1016
post-infection period were the traits used. Additional observations were made on the 1988-born weaners. After the artificial challenge period all animals were drenched with ivermectin at the manufacturer’s recommended dose rate, then returned to graze contaminated pasture. Ram and ewe weaners were held in similar adjoining paddocks. Following 6 weeks of natural challenge, epg and PCVs were recorded, then weaners were drenched with ivermectin and closantel at the recommended dose rates. Larval differentiation of faecal samples bulked within each selection line revealed that the larvae acquired during the natural infection period were almost exclusively H. contorfus. In order to test the resistance of each line to Trichosrrongylus species, all weaners were given an oral infection of 10,000 T. colubriformis larvae I week after ivermectin and closantel treatment. Faecal egg counts were again recorded 4 weeks after this infection. Various strains of H. contortus larvae have been used over the years of the selection experiment. For the data reported here, the strain used for artificial infection originated from a field isolate obtained at the University of New England’s Kirby Research Farm, Armidale. The T. colubriformis strain was provided by the CSIRO McMaster Laboratory, Sydney. Statistical analysis. For these analyses, data were collected from 69 IRH, 47 DRH and 84 CH line weaners. Because of the unbalanced nature of the data, the method of least squares was used for the analysis of variance of epg and PCVs. The least squares means presented are thus free of any bias caused by the unbalanced design. Faecal egg counts were transformed prior to analysis to remove skewness and kurtosis. Effects tested in the analysis of variance were selection line, sex, birth status (twin or single), age of dam, age of infection and first order interactions. With the
exception of line effects, all main effects and interactions found to be not significant were omitted from the final model. A second model was fitted to test the significance of the line x challenge type interaction. For this analysis, the faecal egg counts following each challenge were treated as repeat observations on each animal. The effects of line, animal within line, challenge type and the line x challenge type interaction were then tested using least-squares analysis of variance. In order to remove scale effects when testing the interaction, log-transformed egg counts were expressed in
standard deviation units. RESULTS Line effects were highly significant (P < 0.001) for all traits except PCV of uninfected animals (Table 1). Sex effects were only significant for PCV following natural infection (P
35% 30% 25%
Packed
20%
cell volume
15% 10% 5% 0% <
Uninfected
Artificial challenge
Natural challenge
Selection line
I
IRH
CH
DRH
FIG. 1. Mean packed cell volumes (PCV) in the increased resistance (IRH), decreased resistance (DRH) and control (CH) lines prior to infection, during artificial infection with Haemonchus larvae and after a period of natural infection. All differences were significant (PcO.05) except prior to infection and between the CH and DRH lines after natural infection.
Genetic
of sheep to H. contortus
resistance
TABLE~--SUMMARY OFANALYSES OFVARIANCE FORPACKED CELL VOLUMES (%) IN UNINFECTED
ANIMALS
AND IN THE SAME ANIMALS
FOLLOWINGARTIFICIALANDNATURALCHALLENGEWITH
H.COntOrtus
Source of variation Line Sex Error Packed cell volumes-mean Uninfected Artificial H. contortus Natural H. contortus Faecal egg Artificial Natural Artificial
squares 14.9749 452.9555** 241.8353**
counts-mean squares H. contortus 11.8574** H. contortus 56.7018** T. colubriformis 2.7675**
Degrees of freedom
11.0667 3.6875 96.0395*
10.2239 13.9983 17.2580
0.0006 0.0973 0.0575
0.2423 1.2045 0.1732
1
196
2
Mean squares for faecal egg counts (in log,,, eggs per gram) following three challenge periods are also tabulated. * P
The patterns in faecal egg counts corresponded with the levels of anaemia (Fig. 2). Maximum epg during artificial H. contortus challenge were 2730 in the IRH line (back-transformed), compared with 12,720 in the CH line and 17,400 in the DRH line. Although the comparison of IRH animals with CH and DRH animals was highly significant (P < O.OOl), the
1017
difference between CH and DRH means was not significant. After natural challenge, the corresponding means were 140, 3590 and 8750 epg, with all three means significantly different from each other (PC 0.05). Following artificial challenge with T. colubriformis larvae, the least squares means were 490,840 and 1340 epg, respectively (Fig. 2), again all significantly different from each other (P~0.05). In all cases except initial PCV, the CH line mean was intermediate to those of the divergent selection lines, but there was an interaction between selection line and parasite species (PC 0.01, not tabulated). Following artificial infection with H. contortus, faecal egg counts in the DRH line deviated further from the CH line (- 79%) than the DRH line (+37%). However, following infection with T. colubriformis, the respective deviations were - 42% and + 60%. DISCUSSION
These results demonstrate that selective breeding of Merinos for resistance to artificial challenge with infective H. contortus larvae is effective. Substantial selection progress has been made within a few generations, as was predicted from the estimates of genetic variation in unselected flocks (reviewed by Piper, 1987). They also parallel the findings of Windon, Dineen & Wagland (1987) who have demonstrated genetic differences in Merinos selected on the basis of
_
Artificial
Natural
H. contortus
Artificial
H. contorhrs
T. cdubtiformi~
Selection line I
IRH
CH
1500
4
1
DRH
FIG. 2. Mean worm egg counts in the increased resistance (IRH), decreased resistance (DRH) and control (CH) lines, following artificial and natural H. contortus (Hc) challenge and artificial T. colubriformis (Tc) challenge. All differences were significant (PcO.05) except between CH and DRH following artificial infection with H. contortus.
1018
R. R. WOOLASTON,I. A. BARGERand L. R. PIPER
artificial challenge following vaccination with irradiated T. colubriformis larvae. Baker, Watson & Harvey (1988) also reported progress when selecting Romney sheep for low epg in an environment where Trichostrongylus and Ostertagia were the most prevalent genera. A most encouraging aspect of this study is that the differences in epg between the lines persisted after grazing naturally infected pastures. Simulation studies predict that differences of this magnitude should have a marked effect on requirements for anthelmintic treatment, rate of development of drug resistance and level of pasture contamination (Barger, 1989). The experimental requirements of the selection programme have precluded weaners from each line being run separately, so we have as yet been unable to verify these predictions in the field. The other notable aspect is that selection for resistance to H. contortus has also conferred a degree of resistance to T. colubriformis. On a percentage basis, faecal egg counts in the IRH line deviated less from the CH line following artificial infection with T. colubriformis (42%) than with H. contortus (79%). Surprisingly, the reverse was true for the DRH line (60 and 37%, respectively). This interaction is somewhat puzzling and could be an artifact of either the method used to assess resistance or the method of transformation. However, the latter explanation is unlikely, because similar results were obtained in preliminary analyses using no transformation or a square root transformation. Evidence for a genetic correlation between resistance to the two parasites was also reported by Windon et al. (1987), who found that moderate levels of protection against H. contortus were detectable in the third generation of selection for resistance to T. colubriformis. In another study, Gray & Barger (cited by Barger, 1989) compared the progeny of normal Merino sires with progeny of a sire known to confer a high level of resistance to H. contortus. They found that the lambs genetically resistant to H. contortus also displayed a degree of resistance to natural infections of Trichostrongylus and Ostertagia species. It is possible that the prior exposure of lambs to a high level of H. contortus in this study may have affected the magnitude of response to T. colubriformis, but there is probably also an effect of ageper se. In the study of Gray & Barger (cited above), differences in
resistance between the genotypes disappeared by 8-9 months of age, indicating that the age at which comparisons are made may be critical. It is not clear from our results whether the lines differ primarily in their ability to develop resistance or whether they merely differ in the age at which resistance develops. Nor is it apparent whether the differences in resistance to T. colubriformis were a consequence of the immunity developed to H. contortus or simply another manifestation of the speed at which the lines can mount an immune response. These questions can only by resolved by serial observations made under carefully controlled conditions. Acknowledgements-The significant contributions of Dr L. Le Jambre during the earlier years of this study are gratefully acknowledged. Thanks are also due to A. Lisle, G. Brown, B. Dennison, R. Nethery and G. Uphill for their skilled and enthusiastic technical assistance. Financial assistance for the maintenance of these flocks has been provided by the Australian Wool Corporation. REFERENCES ALBERSG. A. A., GRAYG. D., PIPERL. R., BARKERJ. S. F., LE JAMBREL. F. & BARCER I. A. 1987. The genetics of resistance and resilience to Haemonchus contortus in young Merino sheep. International Journal for Parasitology 17: 1355-1363. BAKERR. L., WATSONT. G. & HARVEYT. G. 1988. Genetic variation in, and selection for, resistance or tolerance to internal nematode parasites in sheep. Proceedings, 3rd World Congress Sheep and Beef Cattle Breeding, INRA, Paris, Vol. I, pp. 637-639. BARGER I. A. 1989. Genetic resistance of hosts and its influence on epidemiology. Veterinary Parasitology 32: 2135. LE JAMBRE L. F. 1978. Host genetic factors in helminth control. In: The Epidemiology and Control of Gastrointestinal Parasites of Sheep in Australia (Edited by DONALD A. D., SOUTHCO~TW. H. and DINEEN J. K.), pp. 137-141. CSIRO, Melbourne. PIPERL. R. 1987. Genetic variation in resistance to internal parasites. In: Merino Improvement Programs in Australia (Edited by MCGU~RK B. J.), pp. 351-363. Australian Wool Corporation, Melbourne. WINDONR. G., DINEENJ. K. & WAGLANDB. M. 1987. Genetic control of immunological responsiveness against the intestinal nematode Trichostrongylus colubrlformis in lambs. In: Merino Improvement Programs in Australia (Edited by MCGUIRK B. J.), pp. 371-375. Australian Wool Corporation, Melbourne.