Some breeds of cattle, and specifically Nguni cattle, have the ability to acquire resistance to tick infestation (Spickett et al. 1989). The Döhne Agricultural Development Institute, which includes the adjacent Campagna Production System, Eastern Cape province, South Africa, has maintained separate herds ofBonsmara and Nguni cattle for several years. Both R. decoloratus and R. microplus are present on these animals. The set-up at the Döhne institute thus presents an excellent opportunity to compare the resistance of Bonsmara and Nguni cattle to natural tick infestations and to explore the dynamics of R. decoloratus and R. microplus, as well as other tick species, on these animals and on the pastures.

Our study had several objectives. Firstly, we aimed to establish the species composition of free-living ticks questing for hosts on the pastures at Döhne and Campagna, and of adult ticks on cattle grazing the pastures at the two localities. Secondly, the study aimed to determine whether the number of questing ticks onthe pastures grazed by Bonsmara cattle and the number of adult ticks on these animals differed from that on the pastures grazed by Nguni cattle and the number of adult ticks on the latter animals. Thirdly, we aimed to determine the seasonality of questing ticks on the pastures and of adult ticks on the cattle.

The study was conducted at the Döhne Agricultural Development Institute (hereafter referred to as Döhne) and at the adjacent Campagna Production System (hereafter referred to as Campagna), both located in the Amahlathi District, Eastern Cape province. The term Döhne complex will be used when referring to both.Döhne (32°31’S, 27°28’E) comprises an area of 992 ha and its vegetation is defined as Döhne Sourveld (Acocks 1988). The dominant grass species are Eragrostis curvula, Eragrostis plana, Sporobolus africanus and Themeda triandra. Campagna (32°29’S, 27°29’E) is 248 ha in extent and its vegetation is characterised by an open, treed savanna with a moderate shrub and grass cover and is classified as Eastern Province Thornveld (Acocks 1988). In addition to the four major grass species atDöhne, the grazing on Campagna consists of Andropogon appendiculatus and Heteropogon contortus.

A herd of about 46 Bonsmara cattle, which included 16 replacement heifers, rotationally grazed five camps of natural veld at Döhne. The camps varied in size(from 12.7 ha to 17.0 ha), with a total surface area of 72 ha. A herd of about 64 Nguni cattle, of which 24 were replacement heifers, similarly grazed five camps of between 9.9 ha and 14.7 ha, with a total surface area of 65.3 ha. At Campagna a herd of approximately 30 Bonsmara cattle, of which seven were replacement heifers, rotationally grazed five camps of between 10.9 ha and 16.2 ha, with a total surface area of 69.4 ha. When grazing was inadequate the animals were fed lucerne hay and winter and summer licks were provided ad libitum. Because of a sudden spate of mortalities amongst the Bonsmara cattle a few years earlier, due probably toBabesia bovis infection, they were treated with the acaricide flumethrin (Drastic Deadline, Bayer Animal Health) at least once a month, administered as a pour-on. The Nguni cattle were treated only when adult ticks became visible.

Ten flannel strips (100 cm x 10 cm) were attached next to each other on a 120-cm-long wooden spar with Velcro tape and the two ends of a twine harness were tied to the tips of the spar. An operator collected ticks questing for hosts from the vegetation by dragging the spar, using the twine harness, across the vegetation over adistance of 100 m (Spickett et al. 1992). Drags were always repeated six times at approximately 50-m intervals in camps grazed by the respective cattle breeds. After each drag, the ticks on the flannel strips were collected using forceps and stored in 70% ethanol in internally labelled glass vials. Drag-sampling of the vegetation was performed monthly between January 2005 and December 2009. With a few exceptions sampling was performed during the fourth week of each month so thatit coincided with the weighing and dipping of the cattle grazing the pastures. Drags were not performed over dew-laden grass early in the morning or immediately after rain, as this would wet the flannel strips and decrease their efficacy for picking up questing ticks. During the five years of the study a total of 1080 drags were made and all the ticks collected were identified, classed by life stage and counted under a stereoscopic microscope. The aggregate numbers of larvae collected permonth were used to illustrate graphically the seasonal abundance of the most numerous tick species.

From February 2007 to January 2010, adult ticks were collected monthly from one side (including half the head and one ear) and from the whole of the upperperineum and tail-brush of three Bonsmara and three Nguni cattle at Döhne and three Bonsmara cattle at Campagna. All the ticks collected were identified, classed by gender, and counted under a stereoscopic microscope. The aggregate numbers of adult ticks collected per month were used to illustrate graphically the seasonality of the most numerous species. Maximum and minimum atmospheric temperatures, rainfall and relative humidity were measured daily at Döhne.

The tick counts were entered into a Microsoft Excel database and all analyses were performed using SAS (SAS Institute Inc. 2006, Edition 9.1.3). Because the data were not normally distributed, owing to a large number of zero counts which makes log transformation in the search for normality unreliable, non-parametric tests(Kruskal–Wallis and Wilcoxon) were performed (Altman 1991) to determine significant differences between groups. The groups comprised ticks collected from Döhne pastures and Bonsmara cattle on these pastures, ticks from Döhne pastures and Nguni cattle on these pastures, and ticks from the pastures and Bonsmara cattle at Campagna. Analyses were performed for all species of questing and parasitic ticks whenever there were adequate numbers. A significance level of α = 0.05 was used for each comparison. Similar methodology was used to compare tick numbers (both questing and parasitic) between years. The Kruskal–Wallis test was conducted to determine general fluctuations. Wilcoxon tests were used for pair-wise year comparisons. To control the overall significance level of α = 0.05 Bonferroni–Holm adjustment for p-values was applied (Holm 1979). Spearman correlation coefficients (Dohoo, Martin & Stryhn 2010) were calculated to determinethe relationship between questing and parasitic tick numbers, and monthly and annual rainfall.

FIGURE 1: (a) Mean monthly maximum and minimum temperatures, (b) monthly rainfall and (c) mean monthly relative humidity at Döhne.

The numbers of larvae of the five major tick species collected annually from the pastures at Döhne and Campagna are summarised in Table 2. During one or more years of the survey the annual total numbers of larvae of each of these species exhibited significant fluctuations (p ≤ 0.05) when compared with the numbers collected in any other year. Only one A. hebraeum larva was recovered during the first three years of the survey, but numbers increased substantially thereafter. The number of R. appendiculatus larvae collected during 2007 was substantially lower than in the previous years, but increased thereafter, whereas its nymphs and the larvae of R. evertsi evertsi exhibited a decreasing trend during the study period. Significantly fewer R. decoloratus larvae were collected during 2006 than in any other year of the survey, whilst slight annual fluctuations in the numbers of R. microplus larvae were observed.

Questing R. appendiculatus larvae reached peak numbers in April, followed by a smaller peak in August. Nymphs peaked in October (Figure 2a). Larvae of R. decoloratus and R. microplus were most abundant during August (Figure 2b), whilst R. evertsi evertsi larvae were most abundant during January, with successivelysmaller peaks in June and November, respectively (Figure 2c). The total numbers of R. evertsi evertsi larvae collected during each year of the survey appeared to be positively but not significantly (p = 0.5) correlated with annual rainfall (Figure 3). It should be noted, however, that this calculation was based on only five observations.

FIGURE 2: Composite seasonal abundance of questing (a) Rhipicephalus appendiculatus larvae and nymphs, (b) Rhipicephalus decoloratus and Rhipicephalus microplus larvae, and (c) Rhipicephalus evertsi evertsi larvae on the pastures at Döhne and Campagna (2005–2009).

FIGURE 3: Annual rainfall and numbers of questing Rhipicephalus evertsi evertsi larvae collected at Döhne and Campagna (2005–2009).

The total numbers of ticks of the seven most abundant species collected annually from the Bonsmara and Nguni cattle at Döhne and the Bonsmara cattle atCampagna are summarised in Table 4. Only two A. hebraeum adults were collected in 2007, but considerably more were collected in 2008 and 2009. The numbers of

H. rufipes and R. microplus fluctuated only slightly (but not significantly) during the three years of the study, whereas significant annual differences were observed in the numbers of the remaining tick species. Moreover, R. evertsi evertsi showed a significant increase and R. follis a significant decrease in numbers. The increasein annual numbers of adult R. evertsi evertsi seemed to be marginally inversely correlated with annual rainfall (Figure 4), with a correlation coefficient of –0.50 for aggregated yearly rainfall and –0.01 for monthly rainfall data. The correlations were, however, not significant. The decrease in annual numbers of R. follis seemed to be marginally positively correlated with annual rainfall (Figure 4), with a correlation coefficient of 0.50 for aggregated yearly rainfall and 0.29 for monthly rainfall.Also these correlations proved to be non-significant.

The seasonal abundances of the six most numerous tick species collected from the cattle at Döhne and Campagna are graphically illustrated in Figure 5. Adult H. rufipes reached peak numbers in midsummer and late summer, whilst R. appendiculatus numbers peaked in spring and late summer (Figure 5a). The numbers of R. decoloratus and R. microplus declined erratically from January to August, increased in September and declined towards November, followed by a marked increase in December (Figure 5b). The greatest numbers of R. evertsi evertsi were present from January to March and in September and the fewest from May to August, whereas most R. follis were collected during August (Figure 5c).

FIGURE 4: Annual rainfall and numbers of adult Rhipicephalus evertsi evertsi and Rhipicephalus follis on cattle at Döhne and Campagna (February 2007 – January 2010).

FIGURE 5: Composite seasonal abundance of adult (a) Hyalomma rufipes and Rhipicephalus appendiculatus, (b) Rhipicephalus decoloratus and Rhipicephalus microplus, and (c) Rhipicephalus evertsi evertsi and Rhipicephalus follis on cattle at Döhne and Campagna (February 2007 – January 2010).

The small numbers of questing ticks collected from the vegetation and parasitic ticks from cattle are probably a consequence of the regular application of acaricides to cattle at Döhne and Campagna. Despite these small numbers, a total of 13 species were represented. In comparison, only six ixodid tick species were collected during 2004 and 2005 from vegetation and cattle at 72 dip-tanks in rural farming areas east of Döhne (Horak et al. 2009). Although the number of ticks collected ateach occasion in the present study was low, aggregate monthly and annual trends could be determined because the survey was conducted over several years.

The immature stages of Haemaphysalis elliptica, R. follis and Rhipicephalus simus all use rodents as hosts (Petney et al. 2004) and H. rufipes use hares and birds (Horak &Fourie 1991; Van Niekerk, Fourie & Horak 2006). Adult H. elliptica use carnivores and large felids as hosts and the adults of the other three ticks use cattle and other large hosts. Engorged female ticks detach from these hosts wherever they may roam, feed or rest, and then seek out a sheltered environment to lay their eggs. Very few larvae of these species are usually collected from the vegetation and in the absence of another explanation it must be assumed that they quest for their smallmammal hosts from the soil surface or from the base of the vegetation.

The long-term (1969–2001) mean annual rainfall at Döhne is 773 mm, although, with the exception of 1992 when only 382 mm was recorded, the 490 mm recorded in 2009 is the lowest for the past 40 years. Estrada-Peña, Horak and Petney (2008) noted that more intense periods of drought in the highveld of Zimbabwe may leadto an expansion of the distribution of A. hebraeum in this region, whereas increased temperatures may have the same consequence in the lowveld. The treed-savanna habitat at Campagna constitutes a suitable vegetational environment for all stages of development of A. hebraeum. Moreover, the lower than normal rainfall in the last year of the study, coupled with sustained high summer temperatures, may have contributed to an optimal environment in which a small existing population of tickscould flourish or in which introduced ticks could become established.

Larvae and adult ticks were collected from the vegetation and cattle at Campagna, but only larvae were collected from the pastures at Döhne. Howell, Walker andNevill (1978) state that A. hebraeum cannot survive in open grasslands, and Horak, Gallivan and Spickett (2011) collected fewest questing A. hebraeum larvae from grasslands compared to woodlands and gullies in a treed savanna locality in the Kruger National Park. The open grass pastures at Döhne would thus seem unlikely to sustain A. hebraeum; nevertheless, some ticks must have completed their life cycles there, hence the presence of questing larvae.

Theiler (1962) stated that R. appendiculatus is consistently absent in open grasslands, whilst Short et al. (1989) reported reduced survival of the free-living stages of R. appendiculatus in short grass habitats in Zimbabwe. This does not appear to be the case at the Döhne complex, where slightly more R. appendiculatus larvae were collected from the open grass pastures and adult ticks from the Bonsmara cattle at Döhne than from the vegetation and cattle in the treed savanna habitat ofCampagna.

R. evertsi evertsi can complete more than one life cycle annually and thus more than one peak in the abundance of immature and adult ticks can occur in a single year. The pattern of summer, winter and spring peaks in the numbers of questing larvae collected from the pastures at Döhne (Figure 5c) is similar to that found forticks on sheep on a highveld farm in the eastern Free State (Horak, Williams & Van Schalkwyk 1991b). The decrease in numbers of questing larvae on the pastures at Döhne with decreasing rainfall, whilst the numbers of adult ticks on cattle increased is difficult to explain.