S. Afr. J. Bot., 1986, 52(2) 189

metres below the surface. They bear fynbos vegetation and High orchid densities and a hybrid the typical include Erica coarctata, WendL, Myrica Satyrium at Blue Downs near quercifolia L., of Muraltia, Struthiola, Hermannia, Zygophyllum and several Restionaceae and Rutaceae. This Blackheath, is not a typical Coastal Fynbos as described by Boucher (1982, 1983) and may be allied to the Elim Lowland Fynbos (Moll L.M. Raitt et al. 1984). Three species of mole-rat (Bathyergidae) occur in the fynbos Department of Botany, University of the , Belville, 7530 Republic of South Africa biome of the south-western Cape, South Africa (Du Toit et al. 1985). Of these the largest, Bathyergus suillus (Schreber), Accepted 30 September 1985 the Cape dune mole and at least one other species occur at Possible reasons for the high densities of Satyrium Blue Downs. Table 1 shows Satyrium distribution and mole­ carneum (Dryand) Sims and S. coriifolium Sw. in dune rat activity in relation to the depth of the soil, i.e. the depth slacks at Blue Downs are considered. The shallow ferricrete of the iron hardpan below the surface. Regression analysis hardpan limiting mole·rat activity is thought to be a primary showed a significant correlation between depth and mole-rat factor, while the high water availability in the dune slacks is probably a contributing cause. The occurrence of putative activity as measured by the number of mole hills per unit area hybrids is reported. (n = 27; r = 0,468; p < 0,02). A highly significant negative correlation was found between the depth of the soil and the Moontlike redes vir hoe digthede van Satyrium carneum occurrence of orchids (n=27; r= -0,753; p < 0,001), and the (Dryand) Sims en S. coriifolium Sw. in duinlaagtes by Blue presence of orchids and mole-rat activity (n= 18; r= -0,718; Downs word in oenskou geneem. Die vlak ferrikreet p 0,001). Since geophytic storage organs are packaged food hardebank wat die aktiwiteit van die knaagdiermolle be perk, < word as 'n primere faktor beskou terwyl die hoe sources, it seems probable that animals affect the density of waterbeskikbaarheid in die duinlaagtes 'n meewerkende geophytes, particularly in the fynbos where geophytes are oorsaak is. Verslag word gelewer oor die voorkoms van abundant (Kruger 1979). There is no doubt that the mole­ veronderstelde hibriede. rats do use geophytes as a food source (Jarvis et al. 1983, 1984; Du Toit et al. 1985). Mole-rats are ubiquitous on the Keywords: Distribution factors, hybrid Satyrium, mole·rats, , and it is likely that the main reason for the high orchids orchid densities attained at Blue Downs is the relative inaccess­ Blue Downs is an area of the Cape Flats close to Blackheath ibility of their tubers. These may be inaccessible owing to near on which an extensive high density sub­ protection by surrounding ferricrete nodules, or because there economic housing scheme is being planned. Part of this area is insufficient soil to burrow in. The Cape dune mole constructs is the habitat of large populations of (Oliver a burrow of some 12-15 cm in diameter (n=5; 13,7 ±0,3 cm) 1977). Of the eleven orchid species occurring there, Satyrium and in soil of 12,6 ± 1,7 cm deep (Table 1) such a burrow carneum (Dryand) Sims, which is considered by Hall & is impossible. Even the smaller species Georychus capensis, Ashton (1983) to be a vulnerable species, and S. corriifolium the Cape mole-rat, with a burrow of a mean diameter of Sw. are the most conspicuous. Colour photographs of these 10 cm (Du Toit et al. 1985) would probably have its tunnels orchids at this site are given by Oliver (1977) and Stewart collapse. et al. (1982: Figures 44, 45). The orchids normally flower in October, which is at the start of the dry season (Figure 1) with peak flowering of S. Table 1 Soil depth, Satyrium density and mole hills carneum occurring about two weeks before that of S. corrii­ (x±SE) folium. Soil depth Orchid density Mole hill density The site has a number of stabilized dunes running through n in em per 0,25 m - 2 per 0,25 m - 2 it, covered with strandveld vegetation similar to Taylor's (1972) Dune 15 > 50 0,9±0,3 Euclea-Rhus Inland Dune Scrub. The dune slacks, where the Slack 12 12,6± 1,7 5,1 ±° 1 ,1 0,1 ±O,I orchids are abundant, have a ferricrete hardpan a few centi-

It has been suggested (Milewski 1984) that some geophytes 16° 525mm are co-adapted with mole-rats which eat, but also potentially disperse and sow the underground storage organ. However, mole-rats do not forage at random and are either able to move from patch to patch of geophytes, or they thoroughly exploit geophyte-rich areas encountered before resuming random foraging (Du Toit et al. 1985). It would therefore seem highly unlikely that high densities of orchids could occur at sites where mole-rats also occur and have ready access to the tubers. In one place where one of the smaller mole-rat species had penetrated the edge of a dune slack, the satyriums were eaten from below. J ASONDJ FMAMJ The sand of the dune slacks was finer and darker than that of the dunes and had a slightly lower pH and a higher electro­ Figure 1 Walter-Leith climate diagram for the Blue Downs area based lyte content (shown by the higher conductivity, Table 2). on D.F. Malan Airport (44 m, temperature) and Eersterivier (30 m, Moisture levels were consistently higher in the slacks, and the rainfall). The orchid site is located between these places. penetrability of the sand was lower. This higher level of 190 S.·Afr. Tydskr. Plantk., 1986, 52(2)

Table 2 Some soil characteristics of Blue Downs (n =3; x ± S.E.) Slack Dune 5 ;--- Water content (070 dry mass) August 1979 15,0 ± 0,2 6,1 ± 0,3 November 1979 9,7 1,3 5,7 0,1 ± ± A B February 1980 3,9 ± 0,2 1,8 ± 0,3 4 ;--- (f) (ij ::J Penetrability' (kg cm - 2) 1J '> November 1979 2,69± 0,14 1,84± 0,18 '6 c 3 r-- February 1980 3,26± 0,30 1,91± 0,15 o Q; .0 pH E ::J KCI 7,81 ± 0,04 8,14± 0,D2 Z 2 Sticky point 7,86± 0,04 8,14± 0,05

Conductivity (~ S cm - 1) 278,7 ±21,0 152,7 ± 16,5

"en = 10)

moisture is probably a secondary factor governing orchid o 26 52 density. The satyriums flower at the start of the dry period I ndex val ues (Figure 1), and in some years when September has been very dry, many of the bolting inflorescences became dehydrated Figure 2 Hybrid index. A - Satyrium corriifolium; B - S. carneum. and died before flowering. Hybrids between S. carneum and S. coriifolium have been It is noted with regret that Oliver's (1977) question 'Orchids reported for this site (Oliver 1977) and elsewhere (Bolus 1918; - or houses' will soon be answered with houses. In the near Duckitt 1976; Stewart et al. 1982). This is not surprising as future the Cape Flats Nature Reserve at the University of the in the large sympatric populations of both species there is Western Cape (Low 1982) comprising but 20 ha may be the considerable overlapping of flowering periods. Hybrids are, only relatively natural vegetation preserved on the Cape Flats, however, rare, and widely scattered, so data for the hybrid some 0,050/0 of the total area. index (Figure 2) had to be collected over several years to have information on six putative hybrids. The hybrid index illu­ Acknowledgement strated is based on thirteen characters (Table 3). Two of these Mr R.O. Moffett is thanked for comments on this paper. properties, inflorescence height and basal leaf length depend perhaps more on the size of the than the other flower References measurements do. It seems that in a year with good rainfall the BOLUS, H. 1918. The orchids of the . Darter, plants grow larger than they otherwise would. Therefore one Cape Town. of the five S. coriifolium plants measured had an index value BOUCHER, C. 1982. Floristic and structural features of the according to Table 3 such that it is in the next interval in the coastal foreland vegetation south of the Berg river, Western histogram, whereas all five S. carneum plants score in the Cape Province. In: Proceedings of a Symposium on Coastal same interval. Because of this size effect and the fact that Lowlands of the Western Cape. ed. Moll, E.J. pp. 21-26. University of the Western Cape, Bellville. hybrids are rare, it is thought that the histogram obtained BOUCHER, C. 1983. Floristic and structural features of the indicates that they are closer to S. coriifolium rather than the coastal foreland vegetation south of the Berg river, Western product of backcrossing. Cape Province, South Africa. Bothalia 14: 669 - 674.

Table 3 Properties used for the Hybrid index

Properties Score

o 2 3 4 1 Spike density (flowers/cm) < 2,0 2,1-2,5 2,6-3,0 3,1-3,5 > 3,6 2 Bract orientation reflexed straight 3 Side sepal orientation reflexed 90° straight 4 Side sepal length (mm) < 14 15 16 17 > 18 5 Lip length (mm) < 15 16-17 18 19-20 > 21 6 Spur length (mm) <11 12-- 13 14-15 16-17 > 18 7 Column length (mm) <10 II 12 13 > 14 8 Stigma lobe width x length 4-5x3-4 4-6x4-6 4x6 2-3x7-8 (mm) (w > I) (w=l) (w < I) 9 Colour yellow-orange salmon white-pink 10 Inflorescence height (cm) < 30 31-36 37-43 44-49 > 50 II Basal leaf orientation (from soil) > 50° 40° - 49° 20° - 39° 10° - 19° < 10° 12 Basal leaf length (cm) < 15 16 17 18 > 19 13 Purple spots on stem base present absent S. Afr. J. Bot., 1986, 52(2) 191

DUCKITT, F. 1976. Orchids of the Western Cape Sandveld area. LOW, A.B. 1982. Nature Reserve on the Cape Flats. Veld & Veld & Flora 62: 16 - 18. Flora 68: 104 - 106. DU TOIT, J.T., JARVIS, J.U.M. & LOUW, G.N. 1985. MILEWSKI, A.V. 1984. A comparison of geophytes and Nutrition and burrowing energetics of the Cape mole-rat underground rodents in Mediterranean Australia and South Georychus capensis. Gecologia 66: 81 - 87. Africa, with special reference to soil nitrogen. In : Proceedings HALL, A.V. & ASHTON, E.R. 1983. Threatened plants of the of the 4th international conference on Mediterranean Cape Peninsula. Threatened Plants Research Group: University Ecosystems, ed. Dell, B. pp. 109 - 110. Botany Department, of Cape Town, . University of Western Australia, Nedlands. JARVIS, J., DAVIES, K. & LOVEGROVE, B. 1983. Mole rat - geophyte interactions. 5th Fynbos Biome Conference, University MOLL, E.J., CAMPBELL, B.M., COWLING, R.M., BOSSI, L., of Cape Town, Rondebosch. JARMAN, M.L. & BOUCHER, c. 1984. A description of JARVIS, J., DAVIES, K. & LOVEGROVE, B. 1984. A major vegetation categories in and adjacent to the Fynbos preliminary investigation of the ecological role of mole-rats in Biome. South African National Scientific Programmes, Report the fynbos of the western Cape (1981 - 1983). In: Terrestrial no. 83. CSIR, Pretoria. ecology in South Africa - project abstracts for 1982-1983, OLIVER, E.G.H. 1977. Orchids - or houses? African Wildlife Committee for Terrestrial Ecosystems pp.65 - 66. South African 31: 10 - 11. National Scientific Programmes. Report no. 86. CSIR, STEWART, J., LINDER, H.P., SCHELPE, E.A. & HALL, Pretoria. A.V. 1982. Wild Orchids of Southern Africa. Macmillan, KRUGER, F.J. 1979. Plant ecology. In: Fynbos ecology: a Johannesburg. preliminary synthesis, eds Day, J., Siegfried, W.R., Louw, G.N. & Jarman, M.L. pp. 88 - 126. South African National TAYLOR, H.C. 1972. Notes on the Vegetation of the Cape Flats. Scientific Programmes Report no. 40. CSIR, Pretoria. Bothalia 10: 637 - 646.