FIGURE 1. Deeply weathered granite with finer grained corestone.


Many vineyards on the undulating lowlands and mountain slopes around Stellenbosch and Helderberg are underlain by granite (Fig. 1) or shale (Fig. 2). Both are strongly weathered. Soils derived from granite and shale might affect grapevines differently through their abilities to supply potassium (K) from their constituent minerals. Where soil potassium (K) levels are adequate, the capacity of the soil minerals to supply potassium (K) to grapevines is probably of no consequence. However, many vineyards receive no potassium (K) until after deficiency symptoms have appeared in order to reduce the risk of luxury potassium (K) consumption leading to high must pH and to poor colour in red wines (Conradie, 1994). Grapevine performance is nevertheless likely to be curtailed long before deficiency symptoms become apparent. The research described here compared the effects of weathered granite and shale-derived lithic soils from previously uncultivated ridge-crest sites on the performance of Merlot grapevines grafted onto two rootstocks of dissimilar parentage. To minimise environmental variables the trial was carried out under controlled conditions.

Materials and methods

Merlot grapevines grafted onto 110 Richter (110 R) (Vitis berlandieri x V. rupestris) and 101-14 Mgt (V. riperia x V. rupestris) rootstocks were grown on granite and shale-derived Glenrosa soils in 1.5 m deep, free-draining, 3.0 m x 1.5 m concrete tanks (Fig. 3). Each soil x rootstock treatment was replicated in six randomised blocks. Lime and superphosphate were mixed into the soils in sufficient quantities to neutralise exchangeable acidity and bring the soil P-concentration to around 30 mg/kg. No potassium (K) was applied. The grapevines were planted during the winter of 2005, trained to a vertical trellis during the 2006/2007 season and pruned to a single two-bud spur per 15 cm cordon length each winter. The soils were separately irrigated by micro sprinklers at matric potentials between -0.07 MPa and -0.08 MPa. Soil sampling and analysis were carried out annually in spring. Wines were made and sensorially assessed in seasons 2007/2008 to 2011/2012. Yield, trunk circumference, cane weight, single leaf area and total leaf area per grapevine were determined each season. Leaves and petioles were sampled at fruit set and analysed. Midday leaf water potentials were determined two days before harvest at soil matric potential c. -0.07 MPa. Midday leaf water potentials averaged -1.55 MPa and did not differ between treatments (mild stress, Myburgh, 2011).


FIGURE 1. Deeply weathered granite with finer grained corestone.
FIGURE 2. Shale profile on Malmesbury Group shale (greywacke). FIGURE 3. Merlot grapevines growing in shale (left) and granite-derived soils (right). Sample text TABLE 1. Characteristics of granite and shale-derived soils in field state and averaged over the trial period (seasons 2007/2008 to 2011/2012). TABLE 2. Effects of soil and rootstock on foliar potassium (K) concentrations, grapevine performance and wine characteristics in Merlot. Data averaged over seasons 2007/08 to 2011/12.

Results and discussion

The granite-derived soil materials contained more stone and coarse sand than the shale (Table 1), corresponding to the generally gravelly character of granite soils. In the field state both soils contained similar amounts of Bray II K (BK). These were below the optimum for the Stellenbosch area of 70 – 80 mg/kg, but above 30 mg/kg, at which level maintenance potassium (K) applications are essential (Conradie, 1994). Averaged over the trial period, BK in the granite soil differed little from its field level, whereas BK in the shale soil was appreciably lower during the trial than in its field state. The granite mineral assemblage therefore maintained BK levels better than that of the shale soils. In their field states the potassium (K) saturations of both soils compared favourably with the recommended saturation of 4% of CEC (Conradie, 1994), but were lower over the trial period, notably in the shale soil. The lower potassium (K) saturations during the trial were due to lime additions that increased the T-values and relative Calcium saturations. In both soils the silt and clay fractions were dominated by, respectively, quartz, which is a primary mineral, and kaolinite, which forms from products of silicate mineral breakdown. Since neither quartz nor kaolinite supply potassium (K), the observed difference in BK probably reflected dissimilarities in the mineral assemblages of the coarse sand fraction and of the finer stone fragments. The coarse material in the granitic soil contained degraded feldspar and a little mica, both of which are potentially capable of supplying small amounts of potassium (K), whereas that of the shale soil mostly consisted of crumbling fragments of rock from which most of the potassium (K) had been removed by weathering (Bühmann et al., 2004).

Average leaf blade potassium (K) concentrations did not differ between soil x rootstock treatments (Table 2) and were within the normal range (0.65% to 1.30%; Conradie, 1994). Neither did the petiole potassium (K) concentrations differ from the norm (1.00% to 2.90%; Conradie, 1994) despite the prevailing low soil potassium (K) saturations. Petiole potassium (K) concentrations were significantly (p ≤ 0.05) higher in the shale x 101-14 Mgt than the granite x 101-14 Mgt and shale x 110 R treatments. Yields were higher in the granite x 110 R than the shale treatments (both rootstocks). However, trunk circumferences were lower in the shale x 110 R than in all other treatments. On both soils cane mass of vines grafted onto 101-14 Mgt were greater than vines grafted onto 110 R. Cane mass of vines grafted onto 110 R were higher on the granite soil than the shale soil. The shale x 110 R treatment was associated with low single-leaf areas, and low total leaf areas per grapevine, particularly in comparison with granite x 101-14 Mgt. The greater cane mass and total leaf areas of grapevines on 101-14 Mgt compared to 110 R (both soils) indicate that 101-14 Mgt promoted a more extensive canopy than 110 R. Overall wine quality from granite x 101-14 Mgt grapevines was significantly higher than that from the shale x 110 R treatment. A contributory factor was the greater fullness in wines from the granite x 101-14 Mgt treatment than in wines produced on 110 R, irrespective of soil.

Merlot/110 R grapevines produced greater yields, trunk circumferences and cane mass on the granite than the shale soil. In contrast these grapevine performance parameters did not differ between soils in the Merlot/101-14 Mgt grapevines. Petiole potassium (K) concentrations were 30% greater in the granite x 110 R than the shale x 110 R treatment. However, this difference was not significant. Conversely, petiole potassium (K) concentrations in grapevines grafted onto 101-14 Mgt were greater on the shale than the granite soil, despite the slightly lower average BK levels and potassium (K) saturations in the shale soil. 110 R may therefore be less able to absorb potassium (K) from low potassium (K) soils than 101-14 Mgt.

Petiole potassium (K) concentrations on the coarser-textured granite soil differed little between rootstocks, but differed between rootstocks on the finer-textured shale soil. The higher petiole potassium (K) concentrations in the shale x 101-14 Mgt treatment were not associated with higher yields or better wine quality than from the shale x 110 R treatment, even though the former treatment resulted in thicker trunks, greater cane masses and greater total leaf areas. Implications of these results are that Merlot/110 R grapevines are likely to produce slightly higher yields from weathered, mostly granite-derived soils than Merlot/101-14 Mgt. Merlot/110 R grapevines may also promote higher yields when grown in granite than in shale soils. Rootstock has little effect on overall wine quality in either soil.


In disturbed, irrigated, limed and P-supplemented soils, Merlot grapevines grafted onto 101-14 Mgt rootstocks growing in a soil derived mainly from granite were more vigorous and produced wine of better quality than those on 110 R growing in a shale-derived soil. Yields tended to be highest in the combination of Merlot/110 R and granite soil.

To the extent that soil texture and mineralogy reflect those of the parent rock, soil and rock type should be considered when selecting rootstocks for Merlot in the western coastal foreland of the Western Cape. Where potassium (K) availability levels are adequate, parent rock type is probably not important from a demarcation viewpoint.


Funding for this trial was provided by Winetech and ARC Infruitec-Nietvoorbij.


Bühmann, C., Nell, J.P. & Samadi, M., 2004. Clay mineral associations in soils formed under Mediterranean-type climate in South Africa. S. Afr. J. Plant Soil 21, 166 – 170.

Conradie, W.J., 1994. Vineyard fertilisation. ARC Fruit, Vine and Wine Research Institute, Nietvoorbij, Private Bag X5026, 7599 Stellenbosch, South Africa.

Myburgh, P.A., 2011. Response of Vitis vinifera L. cv. Merlot to low frequency drip irrigation and partial root zone drying in the Western Cape Coastal Region (Part I): Soil and plant water status. S. Afr. J. Enol. Vitic. 32, 89 – 103.

For further information contact Philip Olivier at

Philip Olivier & John Wooldridge

ARC Infruitec-Nietvoorbij,


You may like to read these:

Go Back