Last update: March 23, 2012 10:54:51 AM E-mail Print


The camp no. 6 veld grazing trial:

An important milestone in the development of pasture research at the Grootfontein College of Agriculture


C H Donaldson

Pasture Research

Grootfontein College of Agriculture

Private Bag X529





Since 1934, various experiments on natural veld have been undertaken in the Karoo Region in an attempt to obtain critical information on the potentialities and reactions of the veld to various treatments. These studies were conducted so that suitable systems of veld management could be devised that would permit maximum utilization of the veld while ensuring its maintenance and/or restoration.

A review of the development of veld research in the Karoo Region from 1934 to 1983 has been submitted by Roux and Vorster (1983). In this review reference is made to the first grazing experiment undertaken at the Grootfontein College of Agriculture. This experiment was initiated by E Parish (principal of the College) in August 1934 with the overall object of determining which grazing system would have the best effect on Karooveld. Apart from minor changes, this experiment is being continued and is also still known as the Camp No 6 grazing experiment. One of the first reports on the progress of this experiment was published in 1937 (Coetzee, 1937). A detailed report of the initial treatments and experimental findings for the period 1934-1950 was submitted by Tidmarsh (1951). The 1964/65 Progress Report on the Camp No 6 experiment was the last detailed presentation of the botanical and animal production results of this trial (Roux, 1965). Certain aspects of the results of this experiment have however been reported by Roux (1966) and Roux (1973).

It was mainly because of the historical significance of the Camp No 6 grazing experiment and its value as the longest running grazing trial in the Republic of South Africa that the writer deemed it necessary during the 75th anniversary of the College to submit data on the long term effect of continuous grazing (at two stocking rates) and simple two- and three-camp fixed seasonal rotational grazing systems on veld condition, animal performance and certain soil parameters.




Details of the initial treatments, experimental layout and modifications made in 1939 have been set out by Tidmarsh (1951). The camps were laid out in the form of parallel rectangular strips (width: length ratio of approximately 1: 10) in the direction of the slope on gently sloping mixed Karoo apron veld. The soils of the area are sandy and are of the Shigalo series. By 1953 it became apparent that the sheep in all the treatments tended to concentrate more on the lower third of the camps because of soil differences. The lower portion of all camps except camp numbers 2 (a) and 2 (b) (continuous grazing treatments) were fenced off and the treatments were continued at the same stocking rate on the upper portion of the camps.

The grazing treatments comprised the following six grazing systems:

System 1;

Six-months rotation. Camps 1 (a) and 1 (b) were grazed alternately for 6 months each by one flock of 12 sheep at a stocking rate of 1 sheep to 1,28 ha for the system. Camp 1 (a) was grazed each year from 14 August to 13 February. Camp 1 (b) was grazed each year from 14 February to 13 August.

System 2;

Continuous grazing. Camp 2 (a) was grazed continuously by one flock of 15 sheep at a stocking rate of 1 sheep to 0,85 ha.

System 3;

Continuous grazing. Camp 2 (b) was grazed continuously by one flock of 10 sheep at a stocking rate of 1 sheep to 1,28 ha.

System 4;

Four-months rotation. Camps 3 (a), 3 (b) and 3 (c) were grazed for 4 months each by one flock of 12 sheep at a stocking rate of 1 sheep to 1,28 ha for the three camp system.

Camp 3 (a) was grazed each year from 14 August to 13 December.

Camp 3 (b) was grazed each year from 14 December to 13 April.

Camp 3 (c) was grazed each year from 14 April to 13 August.

System 5:

One-year rotation. Camps 4 (a) and 4 (b) were grazed alternately for 12 months, each (commencing on 14 August), by one flock of 12 sheep at a stocking rate of 1 sheep to 1,28 ha for this system.

System 6:

Continuous grazing. Camp 5 was grazed continuously by one flock of 8 sheep at a stocking rate of 1 sheep to 1,28 ha.

An exclosure of 3,4 ha of veld protected from all grazing was also provided.


Merino wethers were introduced to the experiment at the "two-tooth" stage and replaced after three to four years. The sheep were weighed at intervals of four weeks after being starved for a period of 18 hours. This practice was abandoned in 1941 and sheep were weighed directly off the veld. The sheep were initially shorn during February but since 1941 shearing was done during August.


Botanical analyses and veld condition

Surveys of the composition and cover of the vegetation were made on several occasions using different methods. Unfortunately no surveys were conducted at the beginning of the trial. The first surveys were made by George Gill, a botanist, in 1942. In these surveys the line intercept method from three separate line transects was used to determine species composition. The plant species of this survey were grouped into 4 groups (Tidmarsh 1951). In an attempt to get some idea of the relative trends in the changes in veld condition over time, it was decided to use the 1942 botanical data and data obtained from a survey conducted during July 1986. The chain survey method of Roux (Tidmarsh and Havenga, 1955) was used for the 1986 surveys. These surveys were conducted on three separate line transects that traversed a camp at fixed intervals. Each line transect was 100m long and the number of strikes were recorded at 1 m intervals. The canopy spread cover, on a plant species basis, was used as the measure of survey. The plant species were grouped in the same groups as those of the 1942 survey. The pioneer grasses and unpalatable bushes were further grouped together, whilst the other grasses and palatable bushes made up the second group and relative index values of 1 and 7 were allotted to the botanical data of these respective groups.


Soil studies

Soil pH (water), nitrogen, phosphorus, potassium, sodium, calcium and magnesium were determined on soil samples obtained from triplicate cores taken at a depth of 50 mm at each of four sampling sites per camp.

The sites were located systematically across the width and in the middle of each camp. The soils were analysed by the soil analysis laboratory at Grootfontein using standard analytical procedures. Soil penetration resistance measurements were made during October 1985 with the rod penetrometer method (Donaldson, 1986). Fifty readings per camp were used.



Veld condition


The results of the veld condition scores of the different grazing treatments in terms of the botanical data of the 1942 and 1986 surveys are presented in Table 1.


The data in Table 1 strongly suggest that the effects of the grazing treatments on veld condition were well manifested within the first 8 years (1934 to 1942) of the experiment, assuming that the veld condition of all the camps were similar at the beginning of the trial. The veld condition scores of the ecologically injudicious treatments, such as grazing from August to February (Camp 1a), continuous grazing at a high stocking rate (Camp 2a) and continuous grazing at a moderately high stocking rate (Camps 2b and 5) were below the average score for the 10 camps during the 1942 surveys. This pattern was still very much in evidence in the veld condition scores of the 1986 data (Table 1) thus indicating that the relatively quick initial effects of the grazing treatments on veld condition set the pattern for future trends. The highly significant (P ≤ 0,01) correlation (r = 0,7) between the veld condition scores of the 1942 and 1986 surveys provide further evidence for the similarity in the veld condition scores between grazing treatments for each of the two respective plant surveys.


Soil characteristics

The long-term effects of different fixed seasonal and yearlong grazing treatments on the chemical and physical characteristics of the top 50 mm soil layer are given in Table 2.



The data in Table 2 indicate that the soil environment between the different camps is highly variable. The statistical comparisons and rankings in Table 2 suggest that more favourable soil changes took place in camps 1 (b), 3 (b) and 6 than in the other camps. There were however no consistent relationships across the eleven treatments between most soil parameters (Table 2) and the veld condition scores of the 1986 surveys (Table 1). Only potassium (K) was significantly correlated (r = 0,76) with veld condition. The main findings of the soil-veld condition relationships are in general agreement with the results of Friedel (1981), namely, that change in plant species composition (the basis of veld condition assessments) did not indicate erosional losses of nutrients from soils of the study sites.


Animal production

The average daily gain (ADG) and the mean annual wool production per sheep of the different grazing systems are summarized in Table 3.




The mean wool production and ADG of sheep for systems 3, 5 and 6 were much higher than those for systems 1, 2 and 4 (Table 3). A feature of the data (Table 3) was the poor performance of the sheep in system 4; these animals lost an average of 2,3 g of live mass per day. Better animal performances were obtained from systems 3, 5 and 6. These live mass and wool production figures are however much lower than the ADG and wool production expected from similar types of sheep on mixed Karooveld. The poor animal performances in this experiment, compared to expected norms, can largely be attributed to a too high stocking rate and the confinement of animals to a camp for fixed periods without considering the amount and quality of the available grazing and adjusting animal numbers accordingly.

The annual live mass gains of sheep for systems 2, 3 and 4 together with mean annual rainfall data over the period 1940/41 to 1965/66 are given in Figure 1.

Annual rainfall and live mass gains of sheep for the periods 1940/41 to 1949/50 and 1958/59 to 1965/66 fluctuated in similar saw-tooth-like patterns (Fig. 1). The fluctuations in rainfall and live mass were however very much less pronounced during the 1949/50 to 1957/58 experimental period. Relatively low levels of animal production were a characteristic feature of the later experimental period (Fig. 1). The sheep of systems 2 and 4 experienced severe losses in live mass during the later half of the 1949/50 to 1957/58 period when the annual rainfalls were generally the same as the long-term mean precipitation. Grasses in the experimental area are, according to Roux (1966), favoured by spring and summer rains, while shrubs are favoured by autumn and winter rains. It would therefore appear that the "grass rains" which were more prevalent from 1950 to 1959 (Roux, 1966) were associated with the constant rainfall pattern illustrated in Fig. 1.

The poor performance of the sheep during the period 1949/50 to 1957/58 was most probably caused by a very low intake of protein which is often experienced when sheep graze mature grassveld during the dormant seasons of the year. An examination of the raw animal data (not presented) for the experimental period 1940/41 to 1965/66 revealed that the very poor performance of the sheep in the three-camp treatment (system 4) was largely due to severe losses in live mass when the animals were confined to Camps 3 (b) and 3 (c) during the grazing periods 14 December to 13 April and 14 April to 13 August respectively. The long term (1940 - 1966) mean daily loss in live mass for the later two grazing periods was 10,9 g and 9,5 g respectively. The sheep gained an average of 14 g per day during the period 14 August to 13 December when they were in Camp 3 (a).



The long term results of the Camp No 6 grazing experiment provide evidence that year-long continuous grazing of mixed Karooveld at high and moderate stocking rates normally leads to poor veld condition and, in the case of the high stocking rate treatment, also to poor animal performance. The yearlong continuous grazing treatments and the one-year two-camp rotation system at a moderate stocking rate resulted in better animal production than the fixed seasonal two-camp and three-camp systems. The relatively poor results of the latter two systems were mainly brought about by the application of injudicious grazing procedures not normally applied in practice.



From a grazing management research point of view the Camp No 6 grazing experiment has long fulfilled its purpose. The results gained from this experiment made possible the synthesis and testing of rotational grazing systems during the forties (Roux & Vorster, 1983). This trial also fulfils a valuable role for demonstration purposes. The initial exploratory work conducted on the soil physico-chemical properties did however indicate that some long term grazing treatments did appear to have influenced certain soil properties. There is however a need for more detailed studies on, for example, the dynamic variability of the soil environment of sites subjected to long term fixed grazing procedures.



COETZEE, P.J.S., 1937. Management and preservation of Karoo veld. Fmg. S. Afr. XII: 287-290. Govt. Printer Pretoria.

DONALDSON, C.H., 1986. Simple techniques for estimating soil compaction. Karoo Agric 3: 55-58.

FRIEDEL, M.H., 1981. Studies of central Australian semi-desert rangelands. II. Range condition and the nutrient dynamics of the herbage layer litter and soil. Aust. J. Bot. 29: 233-245.

ROUX, P.W., 1965. Veld grazing experiment. Camp No 6. Progress Report 1964/65, Director: Karoo Region, Middelburg.

ROUX, P.W., 1966. Die uitwerking van seisoensreënval en beweiding op Gemengde Karooveld. Handl Weidingveren. s. Afr. 1 : 103-110.

ROUX, P.W., 1973. Hoe beweiding met kleinvee Karooveld beïnvloed. Boerderykeur, 115-119, Promedia Publikasies. Braamfontein.

ROUX, P.W. & VORSTER, M., 1983. Development of Veld Management Research in the Karoo Region. Proc. Grassld. Soc. sth. Afr. 18 : 30-34.

TIDMARSH, C.E.M., 1951. Veld Management studies at the Grootfontein College of Agriculture: 1934-1950. Pasture Res. in S. Afr. Progress Report. No 111. Part I. Director: Karoo Region, Middelburg.

TIDMARSH, C.E.M. & HA VENGA, C.M., 1955. The wheel-point method of survey and measurement of semi-open grasslands and Karoo vegetation in S. Africa. Mem. Bot. Surv. S. Afr. No 29. Govt. Printer, Pretoria.



Karoo Agric 3 (8), 1-6