Medical and Veterinary Entomology (1996) 10,3943

Control of Karoo paralysis ticks through vegetation management L . J . F O U R I E , D. J . KOK, L . KRUGEL, A . S N Y M A N and F. VAN D E R L I N G E N Department of Zoology and Entomology, University of the Orange Free State, Bloemfontein, South Africa

Abstract. Karoo paralysis, caused by feeding Zxodes rubicundus females, is a major disease of small stock in South Africa. Control methods currently practised are almost exclusively chemical based. To limit overdependance on chemicals, vegetation manage- ment was investigated as a possible method for control, to be incorporated in an inte- grated tick management system. Laboratory and field experiments were conducted to determine, firstly, the extent of vertical migration and survival of ticks on long and short copper rods which simulated grasses as questing substrates; secondly, the infestation burdens of sheep exposed to similar tick challenges in pens with long and short grass; and thirdly, the effect of trimming the lower crown line of wild olive trees, simulating the browsing effect of goats, on tick density in the immediate environment of the trees. When ticks were exposed to optimal (>45 cm) and sub-optimal (<lo cm) length rods on which to quest, the extent of vertical migration over extended periods of time (up to 87 days) was significantly higher (P < 0.001) for the ticks exposed to long rods. Also, al- most 3 times as many ticks exposed to long rods survived compared to those exposed to short rods. Sheep exposed to long grass were infested by twice as many ticks compared to those exposed to short grass. Tick density at modified wild olive trees ( O h europaea africana) (0.027 ticks/m2) differed significantly (P < 0.05) from that at control trees (0.088 ticks/m2). It is recommended that coarse grazers such as cattle and horses should be used to graze down long grasses before sheep are introduced into camps known to be infested with Z.rubicundus. Similarly, domestic goats can, through their browsing effect on shrubs and trees, modify the vegetation and as such play an important role in an integrated tick management system.

Key words. Karoo paralysis, Zxodes rubicundus, Acari, Ixodidae, control, vegetation management, South Africa.

Introduction

Karoo tick paralysis in South Africa is one of the major small- stock diseases within the distribution range of Ixodes rubicundus Neumann (Acari: Ixodidae). During the period 1983-86 an estimated 115,554 small-stock died due to this disease (Spickett & Heyne, 1988). Various antelope species and cattle are also affected and may suffer mortality (Fourie & Horak, 1987; Spickett & Heyne, 1988; Fourie & Vrahimis, 1989).

The control measures currently practised are almost exclu- sively (97.8%) chemical (Spickett & Heyne, 1988). To limit our overdependence on chemicals for control, various aspects of the biology of Lrubicundus have recently been investigated. These

Correspondence: Professor L. J. Fourie, Department of Zoology and Entomology, P.O. Box 339, Bloemfontein 9300, South Africa.

include its spatial distribution within the habitat (Fourie et al., 1991), life cycle (Fourie & Horak, 1994), host relationships (Fourie et al., 1992; Du Toit et al., 1994). appetence behaviour (Fourie etal., 1993) and vertical migration (Snyman et al., 1994). The purpose of these studies was to expose weak links in the tick's life-cycle and host relationships which can be manipulated to disrupt population growth.

Previously it has been pointed out (Mally, 1904) that the two essential conditions for the occurrence of Karoo paralysis infes- tations are long grass and frost. The author specifically men- tioned that the disease does not occur in short veld. Field observations have shown that most (99%) of the ticks use grasses as questing substrate at a height of c. 45 cm (Fourie et al., 199 I ). These results were confirmed by laboratory experiments which have shown that I.rubicundus preferred long ( A 5 cm) compared to short (10 cm) rods from which to quest (Snyman etal.. 1994).

0 1996 Blackwell Science Ltd 39

30 I J Foitrrr rr rrl

The extent of vertical migration and survival of ticks exposed to suboptiinal and optimal questing substrates for extended periods (>6 weeks) has not yet been investigated.

K a r o o paralysis is endemic to the southern Orange Free State, an area of the central highveld of South Africa which is mainly used f o r sheep farming. This area is characterized by grassland and flattened hills and ridges covered with lush woody vegeta- tion (Acock5. 1988). Ixodr.~ ruhicundu.~ ticks are confined to the hilly area\ where they occur i n close association with shrub or tree apecies characterized by a dense crown cover with a mat of decaying leaf litter underneath (Stampa. 1959; Fourieerul., 1991). 68'7 of' ihe questing ticks collected during visual sampling in this area occurred under the perimeter of the crowns of, or close to. nild olive trees (Olecr europuen ufricunu). or associations consi5ting ofoliize trees together with other shrubs (Fourie etal. . 1991 1. The wild olive tree is also a preferred food plant of do- mestic goats (Kok et d.. 1994) which utilize the trees in such a way thar they may disturb the tick's favourable microhabitat Lvhich could lead to a disruption of the life-cycle and higher inortaliiies in the aphagous tick populations.

The hipothesi.; set is that vegetation management will reduce vertical migration and survival of I. rubicundus and decrease the probability of host-tick contact. The specific objectives of this stud) were. first. to determine the extent of vertical migration and s u n ival of ticks exposed to normal length and suboptimal length questing wbstrates: second. to determine the infestation rate of hosts with I.nrbicundus ticks questing from normal as opposed to sub-optimal length grasses: and third, to determine the effect of wild olive tree modification (trimming the lower crown line to simulate the browsing effect of goats) on tick populations a\sociated with these trees.

Material and Methods

bi,rrii-tr/ ,ni,ynition. Experiments were conducted on adult 1. t i t / k w i d u \ teniales obtained from nymphs fed on their natural ho\t (Fourie P I LII . . 1992). the rock elephant shrew (Elephantulus m i r r u \ ! E n p g e d nymphs were kept in the dark at 25°C and r.h. > W i until the! moulted into adults. Two groups of sixty 2- 3-M eck-old female ticks were placed into similarly sized (.;I x 3: x 6 mi) metal arenas each fitted with a wooden base. Each itrciia contained 256 equally spaced vertical copper rods (0 .15 ciii diameter) S cm and 30 cm in length respectively iSn>maii ~ 1 . . 1994). Ticks were prevented from escaping by the u w r)f an electric barrier (Browning, 1976). The two metal arena5 were placed into a closed glass container (100 x SO x SO cm) i n a rooni in which the temperature and light regime could be controlled. The humidity within the container was kept at r .h . >XOC by using murated salt solutions (Winston & Bates. iW0) plitced i n wide-mouthed. 1 litre plastic bottles as well as vertnictilite nioistcned with water. which covered the bases of the @a\\ container and arenas. Experiments were conducted at 30 L i 'C' and il light regime of IOL: I4D. Over a period of 48 d a y the number of females migrating vertically (observed on the rod., in each ofthe two arenas were recorded three times per da! fmorning. noon and afternoon). At the termination of the trial the number of' living and dead ticks were also recorded. The

experiment was repeated with &&week old female ticks ( n = 30) monitored over a period of 87 days.

Infestation rutes of hosts. The rate of infestation of sheep with 1.rubicundus was determined in two pens (13 x 13 m) 4 m apart and erected on a pasture of natural grass not infested with Z.rubicundus. Prior to fencing, the grass in one of the pens was mowed to a uniform length of about 10 cni, whereas the other area was left intact and consisted of grass blades and inflorescences of which some extended as high as 60 cm. Prior to seeding the pens with ticks, both areas were thoroughly wetted with the aid of a sprinkler system, to ensure a favourable humidity for the ticks.

Ticks (106 male and 94 female; field collected as well as labo- ratory bred) were distributed randomly in each of the pens 2 m inward from each border. Three days were allowed for vertical migration following which two Merino sheep of the same sex and similar size were placed into each pen. Female Z.rubicundus may take 4-7 days to engorge (Stampa, 1959) and sheep were therefore exposed for only 3 days before they were thoroughly examined for the presence of attached and unattached ticks, which were counted and removed. The pens were again thoroughly wetted and seeded with additional male and female ticks to keep the tick challenge similar in the two pens. The same procedure as above was followed except for the fact that the sheep were switched. The experiment was repeated for a third time with the same sheep, switched around again.

During analysis of the data, no distinction was made between tick sex or attached and unattached ticks, because the purpose of the experiments was to determine pick-up rate. The results of the three experiments were pooled.

Wild olive tree modification. This study was conducted on the farm Preezfontein in the south-western Orange Free State. The False Upper Karoo vegetation and the climate of the area have previously been described by Fourie et ul. (1989, 1991).

Twenty wild olive trees occurring on the southern slopes of hills and ridges in the study locality (c. 190 ha fenced area of natural veld) were marked. Ten of these trees, selected at ran- dom, were subjected to mechanical modification. Branches slop- ing down were clipped up to a height of 1.5 m from the ground (Fig. la). Undergrowth was also removed. This procedure was conducted during 1990. During the subsequent 2 years, clearing of undergrowth and pruning of branches took place on a regular basis. Control trees were left intact (Fig. 1 b).

From April to August 1993, a period spanning the activity period of adult I.rubicundus, ticks were sampled at the ten modified and the ten control trees at approximately 14-day intervals. Sampling was always between 08.00 and 1O.W hours. A dragging technique, with cotton blankets (70 x 96 cm), was used to collect ticks from the grass. Two persons, standing next to each other and immediately outside the perimeter of the crown cover of the trees, simultaneously dragged around each tree. This procedure was adopted to ensure that an approximately 2 m wide area could be sampled. Most (99%) questing 1.rubicundus ticks occur on the perimeter or within 2 m of the crown cover of wild olive trees (Fourie et al., 1991). After 3 m of dragging, the collected ticks were removed from the blankets and placed in bottles. Dragging subsequently continued until the whole circumference of the tree was covered.

I996 Blackwell Science Ltd. Mrtfictrl nnd Vererincwy En/omolog,v 10: 39-43

Control of Karoo paralysis ricks 41

b m \

Fig. 1 . Naturally occurring (a: above) and mechanically modified (b: below) wild olive trees ( O h europaea africana).

The total number of ticks collected at each of the trees were expressed in terms of density (ticks/m*) and a nonparametric (Mann-Whitney) test was used to compare values.

Results

Vertical migration

Except for day 1 during the first experiment and day 9 for the second experiment, significantly more (unpaired t test; P < 0.0001) ticks migrated vertically on a daily basis on the long compared to the short rods during both of the experiments. The extent of vertical migration declined gradually after maximum levels had been reached (Fig. 2). At the termination of the two experiments 40% and 33% respectively of the ticks exposed to the long rods was still alive compared to the 15% and 1 1 % of the ticks exposed to the short rods.

Infestation rates of hosts

A summary (pooled data) of the numbers of ticks collected from the sheep confined to the pens with long and short grass is given in Table 1; twice as many ticks were collected from sheep

Vertical migration (%) 80

8 Long rod8 Short rods

60

40

Table 1. Total numbers of ticks collected from Merino sheep, after 3 days of exposure to tick infested pens with long and short grass.

Attached Unattached

Male Female Male Female Total

Long grass 0 55 10 3 68 Shortgrass 0 28 4 0 32

in the pens with long grass. The pick-up rate of sheep in the long grass was 3.78 f 0.57 tickslsheeplday compared to 1.78 k 0.54 ticks/sheep/day in the short grass. These values differed signifi- cantly ( t = 4.6; P = 0.006).

Wild olive tree modiJcation

A summary of statistics on the results obtained with dragging at modified and control wild olive trees is given in Table 2. Significantly fewer (P = 0.023) ticks were collected at modified compared to control trees. At seven of the modified trees no ticks were collected, compared to two of the control trees.

Table 2. Comparative statistics on tick density (tickslm') at modified and control wild olive trees.

- x SE min max n

Modified trees 0.027 0.02 0 0.201 10 Control trees 0.088 0.026 0 0.236 10

Discussion

A comprehensive discussion concerning vertical migration of 1.rubicundus and selection of substrates on which to quest has been given by Snyman et al. (1994) and will not be reiterated.

0 1996 Blackwell Science Ltd, Medical and Veterinary Entomology 10: 39-43

32 I.. J. fiiurrc et ul.

l h e prexnt study has, however, demonstrated that both the ex- tent of' vertical migration and survival are decreased i f I.ruhicur2du.s ticks are exposed to suboptimal length substrates on which to quest. Since all the laboratory experiments were conducted under similar temperature and relative humidity con- ditions, these factors could not be responsible for the observed difference in tick mortality. A plausible explanation is that the extent of activity between the ticks exposed to the long and short rods differed. Suboptimal questing substrates may induce ticks to move more in search of optimal length substrates on which to quest. thereby depleting their energy reserves more rapidly, resulting in increased mortality. This contention remains to be verified.

Based on the results obtained in the laboratory, the outcome of the field experiments on the rate at which sheep, exposed to short and long grass, become infested with ticks, was to be expected. It was. however, envisaged that the sheep confined to the short grass pen would have had lower than the observed infestation burdens. The high density of ticks within the pens ( 1.2-1.7 ticksh'), and the fact that I.rubicundus are attracted to a CO. source over short (53 m) distances (Fourie et al., 1993). may explain the observed results. Sheep most probably were infested whilst lying down.

The pest management concept crnphasizes selectiveness in control. fitting control methods into the biology of the target spe- cies and the social and economic framework of the problem (Geier, 1966). Experimental data gained during the present study indicate that when grasses are grazed down both vertical migra- tion and the probability of tick-host contact will be decreased. Manipulation of the grass length can be achieved without mon- etary investment and from a grazing perspective it is quite feasi- ble. Ixodes rubicundus in the southern Orange Free State commences seasonal activity during autumn and reaches a peak in activity during the winter (Fourie et nl., 1989). periods during which grasses show little or no growth. Vegetation on hilly areas is more protected from extreme temperatures and wind, com- pared to the plains, and provide a higher quality forage during the winter (De Waal. 1986). So livestock prefer the hilly areas where they also come into contact with I.rubicundus (Fourie & Kok. 1992). Since many of the sheep farmers also have a cattle component it is recommended that these animals, along with other coarse grazers such as horses, should be used to graze down the grasses before sheep are introduced into camps that are known to be infested with 1.ruhicundus. The cattle may also reduce tick numbers by eating a significant number of questing ticks while grazing on grasses (Ito. 1990).

Various tick species are known to be closely associated with certain vegetation types or particular plant species (Wilkinson, 1067; Fourie et ui.. 1991; Adler et a ] . , 1992; Stafford & Magnarelli, 1993). In the present study significantly fewer ticks were sampled at modified compared to normal unmodified wild olive trees. This difference may be due to the fact that undis- turbed foliage and leaf litter of the trees protect the free-living stages against desiccation, thereby enhancing survival. Rock elephant shrews, natural host to the immature stages of /.rrthic.rmdus, have been shown to be closely associated with wild olive trees (Du Toit. Fourie and Du Preez, unpublished data). Unmodified wild dive trees may also attract more rock elephant

shrews, which utilize the edible fruits, use the dense leaf canopy as shelter, or feed on the insects in the leaf litter.

Vegetation management for the control of ticks is divided into different categories, namely biological, which includes plants inimical to survival of non-parasitic stages of the tick, e.g. Stylosanthes spp. (Sutherst et al., 1982), physical, e.g. mechani- cal clearing and burning (Barnard, 1986; Presley & Hair, 1988) and chemical, e.g. herbicides (Wilkinson, 1977; Clymer et al., 1970). Domestic stock which can alter the structure of the veg- etation could be considered as another biological method. Do- mestic goats affect the structure of shrubs and trees because they are mainly browsers (Maher, 1945; Campbell et al., 1962; Du Toit, 1972; Grunow, 1980; Warren et al., 1984) in comparison to sheep and cattle which are mainly grazers. The wild olive tree is a preferred food plant of domestic goats in the south-western Orange Free State (Kok et al., 1994). Domestic goats can there- fore modify the vegetation through their browsing effect on shrubs and trees, and as such play an important role in an integrated tick control system. By including domestic goats in the foraging sys- tem a more discreet utilization of the available vegetation can be accomplished compared to that of sheep alone. As such, an addi- tional source of income can be realized. Plants which otherwise would have been left unutilized are now converted into a commercial system of animal production.

The chances of an animal contracting Karoo paralysis is re- lated to its infestation density (Fourie et al., 1989, 1992). Any approach which could reduce the probability of tick-host con- tact will therefore decrease the chances of an animal contracting the disease (Fourie & Kok, 1992). We believe that the findings of this study form a rational basis for the formulation and imple- mentation of an integrated tick management system against I. rubicundus in the south-westem Orange Free State.

Acknowledgments

We thank Mr J. van Niekerk for the use of his farm Preezfontein. The technical assistance of Messrs P. Mayegiso, J. Mayegiso and E. Williams is gratefully acknowledged. The research was funded by the FRD, Meat Board and Hoechst Ag-Vet.

References

Acocks, J.P.H. (1988) Veld types of South Africa with accompanying veld type map, 3rd edn. Memoirs of the Botanical Survey of South Africa, 57, 146pp.

Adler, G.H.,Telford, S.K., 111, Wilson, M.L. & Spielman,A. (1992)Veg- etation structure influences the burden of immature Ixodes dammini on its main host, Peromyscus leucopus. Parasitology, 105, 105-1 10.

Barnard, D.R. (1986) Density perturbation in populations of Amblyomma umericanum (Acari: Ixodidae) in beef cattle forage areas in response to two regimens of vegetation management. Journal of Economic Entomology, 79, 122-127.

Browning, T.O. ( 1 976) The aggregation of questing ticks, Rhipicephalus pulchetlus on grass stems, with observations on R.appendiculatus. Physiological Entomology$ 1, 107-1 14.

Campbell. Q.P., Ebersohn, EP. & Von Broembsen, H.H. (1962) Browsing by goats and its effects on the vegetation. Herbage Abstracts, 32,273-275.

0 1996 Blackwell Science Ltd. Medical and Veterinary Entomology 10: 39-43

Control of Karoo paralysis ticks 43

Clymer, B.C., Howell, D.E. & Haw, J.A. (1970) Environmental altera- tion in recreational areas by mechanical and chemical treatment as a means of lone star tick control. Journal of Economic Entomology, 63, 504-509.

De Waal, H.O. (1986) Die voedingswaarde van veldweiding van die Sentrale Oranje Vrystaat vir lakterende skape met spesiale verwysing na die rol van aanvullende energie en ruprote'ien. (English abstract) Ph.D. thesis, University of Stellenbosch.

Du Toit, P.F. (1972) The goat in a bush-grass community. Proceedings of the Grassland Society of South Africa, 7 , 6 5 0 .

Du Toit, J.S., Fourie, L.J. & Horak, I.G. (1994) Sequential feeding of Ixodes rubicundus on their natural host Elephantulus myurus: effects on engorgement and moulting success. Onderstepoort Journal of Veterinary Research, 61,. 143-147.

Fourie, L.J. & Horak, I.G. (1987) Tick-induced paralysis of springbok. South African Journal of Wildlife Research, 17, 131-133.

Fourie, L.J. & Horak, I.G. (1994) The life cycle of Ixodes rubicundus (Acari: Ixodidae) and its adaptations to a hot, dry environment. Experimental and Applied Acarology, 18.23-35.

Fourie, W., Horak, I.G. &Van den Heever, J.J. (1992) The relative host status of rock elephant shrews Elephantulus myurus and Namaqua rock mice Aethomys namaquensis for economically important ticks. South African Journal of Zoology, 27, 108-1 14.

Fourie, L.J. & Kok, O.B. (1992) The role of host behaviour in tick-host interactions: a domestic host-paralysis tick model. Experimental and Applied Acarology, 13,213-225.

Fourie, L.J., Kok, O.B. & Van Zyl, J.M. (1991) Spatial distribution of the Karoo paralysis tick Ixodes rubicundus (Acari: Ixodidae) withhin a false upper Karoo veld type. Experimental and Applied Acarology, 11,3749.

Fourie, L.J., Petney, T.N., Horak, I.G. & De Jager, C. (1989) Seasonal incidence of Karoo paralysis in relation to the infestation density of female Ixodes rubicundus. Veterinary Parasitology, 33, 319-328.

Fourie, L.J., Snyman, A,, Kok, D.J., Horak, I.G. &Van Zyl, J.M. (1993) The appetence behaviour of two South African paralysis-inducing ixodid ticks. Experimental and Applied Acarology, 17,921-930.

Fourie, L.J. & Vrahimis, S. (1989) Tick-induced paralysis and mortality of gemsbok. South African Journal of Wildlife Research, 19,118-121.

Geier, P.W. (1966) Management of insect pests. Annual Review of Entomology, 11,471-490.

Grunow, J.O. (1980) Feed and habitat preferences among some large herbivores on African veld. Proceedings of the Grassland Society of South Africa, 15, 141-146.

Ito, I. (1990) Managed grassland in Japan. Ecosystems of the World 17A (ed. by A. I. Breymeyer), pp. 129-147. Elsevier, Amsterdam.

Kok, O.B., Fourie, L.J. & Barkhuizen, L.M. (1994) Diet of free ranging Angora goats in a False Upper Karoo veld type. South African Journal ofAnimal Science, 24,80-86.

Maher, C. (1945) The goat: friend of foe? Easr African Agricultural Journal, 11, 115-121.

Mally, C.W. (1904) Notes on the so-called paralysis tick, Ixodespilosus. Agricultural Journal of the Cape of Good Hope, 25,291-296.

Presley, S.M. & Hair, J.A. (1988) Lone star tick (Acari: Ixodidae) mangement by host manipulation through habitat modification. Journal of Medical Entomology, 2578-8 1.

Snyman, A,, Fourie, L.J., Kok, D.J. & Horak, I.G. (1994) Vertical migration of adult Ixodes rubicundus the Karoo paralysis tick (Acari: Ixodidae). Experimental andAppliedAcarology, 18,101-1 10.

Spickett, A.M. & Heyne, H. (1988)A survey of Karoo tick paralysis in South Africa. Onderstepoort Journal of Veterinary Research, 55.89-92.

Stafford, K.C., I11 and Magnarelli, L.A. (1993) Spatial and temporal patterns of lxodes scapulan's (Acari: Ixodidae) in Southeastern Connecticut. Journal of Medical Entomology, 30,762-77 1.

Stampa, S. (1959) Tick paralysis in the Karoo areas of South Africa. Onderstepoorr Journal of Veterinary Research, 28, 169-227.

Sutherst, R.W., Jones, R.J. & Schnitzerling, H.J. (1982) Tropical legume kills cattle ticks. Nature, 295, 320-321.

Warren, L.E., Ueckert, D.N. & Shelton, J.M. (1984) Comparative diets of Rambouillet, Bardado and Karakul sheep and Spanish and Angora goats. Journal of Range Management, 37, 172-180.

Wilkinson, P.R. (1967) The distribution of Dermacenror ticks in Canada in relation to bioclimatic zones. Canadian Journal of Zoology, 45, 517-537.

Wilkinson, P.R. (1977) Effect of herbicidal killing of shrubs on abun- dance of adult Dermacentor andersoni (Acarina: Ixodidae) in British Columbia. Journal of Medical Entomology, 13, 71 3-718.

Winston, P.W. & Bates, D.H. (1960) Saturated solutions for the control of humidity in biological research. Ecology, 41,232-237.

Accepted 13 March 1995

0 1996 Blackwell Science Ltd, Medical and Veterinary Entomology 10: 39-43