Monthly Archives: August 2023

Nature of the Interaction dynamics in the Tanqua Karoo

Geoff Tribe, A. David Marais & Karin Sternberg

Environmental impact assessments

An Environmental Impact Assessment (EIA) is nowadays required before any development can proceed – yet how accurate are they? The year in question, the season, the size and duration of the survey (weeks/months/years) can influence the outcome of the EIA. What would constitute a minimum duration for such a survey? An ecological research project would require a minimum of 3 years, where 5 years is passable, but 10 years would give far better insight. EIAs are often inadequate in terms of genuine environmental protection, and many species are missed, particularly the geophytes and ephemerals, the invertebrates, frogs, birds and bats. Some EIAs are designed to facilitate and even legitimize development where it should not happen. They are mere snapshots in time and miss the far greater, fundamental ecological processes and responses. An erroneous decision could have far-reaching consequences for very sensitive and less well researched areas. The flora and fauna must be researched well, especially in areas that had not previously been well documented. 

With sedentary plants and perennials which can be more readily located, identified and enumerated, a reliable assessment is more easily accomplished. Yet seeds from some plants can stay dormant for decades until favourable conditions cause them to germinate. This applies also to the erratic flowering of underground bulbous plants triggered usually by substantial rains. An example of this in relation to the account below is the dubbeltjiedoring, Tribulus terrestris, which prior to the flood was rare on the farm. Following the exceptional rains the unwelcome thorns appeared en masse especially around the disused sheep pen and along the banks of the once dry streams, forming dense mats with single plants often covering an expanse of one metre or more in diameter and leaving literally thousands of the thorns to germinate when conditions are again favourable.

When it comes to mobile organisms such as birds, animals and insects which can migrate from regions when adverse conditions persist, re-colonization is necessary when conditions improve. This would necessitate refugesfor such organisms where, within the vast area affected, there are atypical areas that remain favourable due perhaps to underground water or seepages. Would such ‘refuges’ be identified during EIA? Re-colonization can be accomplished by migration or in the case of plants by seeds being blown or carried by animals back into the affected area. This would only be possible if the EIA properly takes into account the region and mitigation of damage and risk is done by judicious sparing of some zones for development.  

The cyclical effect on nature through seasonal and climatic fluctuations determines and requires adaptive responses by different organisms. Yet it is surprising to see how readily reactions can occur in response to changes in the status quo. An example of this was the changes that occurred on a farm in the Tanqua Karoo about 35 km north-east of the town of Touwsrivier. Some interesting observations on this farm will be described after an account of recent climatic conditions. 

Tanqua Karoo

Fig. 1. The succulent Karoo scrub on the farm with the shearing shed in the far distance.

The karroid vegetation of the Tanqua Karoo is composed of xerophytic, semi-desert shrubland with a large number of succulent-leaved species [Fig 1]. There is no surface water on the farm. A wind-pump supplies ground-water when required.  From 20 years ago, average rainfalls were recorded on this farm. Although one year was never the same as the year before, the fluctuations in the fauna and flora were not drastic. However, a drought persisted for the preceding eight years during which time a marked deterioration of the veld was observed. Many succulent plants died and others failed to flower. Many insect species either disappeared or became rare. Four colonies of Cape honeybees (Apis mellifera capensis) two of which nested in aardvark burrows [Fig. 2] and two in clefts in shale outcrops [Fig.3], absconded. The annual rainfall over the past decade averaged 102mm with a range of 34 to 226mm. With the exception of 2018, the preceding 5 years had rainfalls of less than 90mm per year. Generally, the rainfall during November and December was <10% of the annual rainfall but in 2021 it was 26mm (37%).  It was completely different in December 2022.

Fig. 2. Honeybee nest in a deserted aardvark burrow which was pestered by banded bee pirates.

Fig. 3. Honeybee nest in a crevice in a shale outcrop on the farm.

The drought was broken by recurrent episodes of significant rainfall from November 2022 to May 2023 which were unprecedented for the last 20 years. The rainfall of 143mm in December resulted in a flood. Indeed, the Tanqua Karoo and adjacent regions experienced widespread floods. Dry river streams turned into raging torrents and the sand was washed down from the hills and deposited onto the plain below, leaving a wide river bed with solid shale bedrock. Over the six months of increased rainfall, the growth of the vygies in particular was particularly good. Whereas one could easily walk in the empty patches between the vygies, they now coalesced into an aggregation as they increased in growth. Vachellia karroo  (soetdoring “ or karoodoring) immediately responds to substantial rainfall .  Vachellia karroo flowers develop only on the young growth of that season and so growth, dependent on rainfall, precedes flowering by 4 to 5 weeks. Within a short while these trees flowered twice in succession and became an attraction for all manner of insects including beetles, wasps, flies, caterpillars, butterflies, bugs, aphids, and solitary bees … but no honeybees were seen. Presumably the Cape honeybees that were observed collecting water at various seepages throughout the farm were visiting alternative sources which were more profitable.

Fig. 4. The succulent Tylecodon paniculatus which grows on the surrounding hills.

Due to the floods, the Vachellia karroo and Searsia burchellii trees which were scattered along the rivulets had their roots exposed.  The roots had been unable to penetrate the shale bedrock but instead had followed along the banks of the river, some being as long as perhaps 12 metres. In most cases it appeared that the combined mass of the roots of a tree was greater than that part above the soil. Another observation was that many of the larger succulent species such as Tylecodon paniculatus [Fig. 4] and Tylecodon wallichii had rotted and collapsed due to the perpetually rain- soaked soil. Surprisingly, larvae of an unidentified longhorn beetle (Cerambycidae) which are usually found in the solid wood of dead trees were found in the soggy, water-laden stems of a T. paniculatus [Fig.5].

Fig. 5. Longhorn beetle larvae in the stem of a rotting Tylecodon paniculatus.


Although five species of snakes and several lizard species occur on the farm, until this flooding no frogs had ever been seen. Yet over a period of at least six months the newly filled hollows in the impermeable shale of some of the riverbeds by seepages contained many hundreds of tadpoles. Two frog species were identified.  The Karoo toad (Bufo gariepensis, also named Vandijkophrynus gariepensis)) [Fig.6] can, under favourable conditions, reproduce in large numbers because as many as 20 000 eggs can be deposited by a single female.  These distinctive eggs are united into strings [Fig.7] by copious amounts of jelly-like oviducal secretion. The eggs hatch into small, dark tadpoles which concentrate into tight free-swimming clusters [Fig.8]. The adult Karoo toad may vary considerably in colouration but is cryptic in its prevailing surroundings.

Fig. 6. The Karoo toad, Bufo gariepensis.

Fig. 7. A string of Bufo gariepensis eggs.

Fig. 8. The aggregating tadpoles of Bufo gariepensis.

The second species, the Common Platanna or African clawed toad (Xenopus laevis) [Fig.9] has webbed toes, and lacks a tympanum, tongue and movable eyelids as an adaptation for an aquatic life. Adults are both predators and scavengers. Their eggs are small, heavily pigmented and are enclosed in individual jelly capsules attached to submerged objects. 

Fig. 9. The aquatic Platana, Xenopus laevis.

The question arises from whence these amphibians came? Because both species are indigenous over this vast and arid region, it can be surmised that they had buried themselves in the sandy banks of the dry streamlets in order to survive the adverse drought conditions. This phenomenon has been observed in the Namib Desert where frogs were found a metre deep in sand under a dried pool in dry-hibernation. Had the Tanqua floods not brought them to activity, they would not have been known to occur on the farm.

Honeybee races at hybrid zone

The Cape honeybee (Apis mellifera capensis) [Fig.10] is restricted to the winter rainfall region of southern Africa but is purported to have an interface with the Savannah bee (Apis mellifera scutellata) along the margins of this region which has been designated as the hybrid zone.  Because the outer margins where the two honeybee races converge are mostly semi-desert, it has been postulated that the low numbers of wild colonies and the correspondingly reduced population pressure between the two sub-species, could result in co-existence but changes in the winter and summer rain could influence the success of  a given species occupying the zone.

Fig. 10. The Cape honeybee, Apis mellifera capensis, the dominant sub-species on the farm.

Fig. 11. Apis mellifera scutellata visiting the flowers of Eberlanzia ferox.

Several other factors may also influence the habitation and cohabitation of the bees. The area where the highest numbers of ovarioles occur in laying-workers of the Cape bee is regarded as the heartof the capensis distribution.  This lies in an inverted triangle that can be drawn from Stellenbosch in the west, to Swellendam in the east, and to Cape Agulhas in the south. The Cape bee has maintained its dominance in the winter rainfall regions for many thousands of years. The predominance of capensis is also observed when scutellata hives are brought into their region. Within a few years such hives become occupied by capensis because capensis workers find their way into scutellata colonies and become laying-workers and eventually take over these colonies by becoming pseudo-queens. This is due to the higher level of queen pheromone possessed by the Cape bees.

The ‘Capensis Calamity’

Cape honeybees taken to the aloe-flow on the Springbokvlaktes north of Pretoria were able to infiltrate scutellata colonies during the manipulations which take place when ‘making increase’.  This is accomplished by taking advantage of the massive amounts of pollen available by dividing a hive and allowing queens to be produced in the queenless brood placed above the queen excluder. Such conditions are ideal for the acceptance of a Cape bee into the queenless partition and here to rapidly develop into a pseudo-queen. The upper partition rapidly fills up with capensis brood and then Cape honey bees emerge.

Despite this proliferation, it appears that capensis colonies do not survive for long in the summer rainfall region but steadily dwindle in numbers and eventually die out. Cape bees rarely are able to invade healthy widely scattered natural colonies of scutellata, but will readily invade commercial colonies because of their proximity in apiaries where regular manipulation takes place. There is much anecdotal evidence that honeybees are adapted to certain environments where they flourish, yet merely maintain themselves when taken to a different environment. For instance, honeybee colonies hived in the Delmas area and brought to Pretoria where they were average producers compared to resident colonies, excelled when taken back to the sunflowers around Delmas each year. Cape bees in Pretoria were seen in greater numbers to visit Cape plants like vygies compared to scutellata bees. After more than 30 years in scutellata territory, other than as invaders of manipulated commercial scutellata hives, the Cape bee has failed to become established in the summer rainfall region.

Shifting boundary

In May 2023 on one of the hills on the Tanqua farm where the vygie Eberlanzia ferox was flowering, some bright yellow honeybees were found with little pubescence on their abdomens which could only be scutellata [Fig.11]. Had they temporarily expanded into capensis territory due to the excessive (summer) rains and the resulting mass flowering of plants which could facilitate such an expansion? Although the identity of the bees has not been determined scientifically by dissection, having worked with both races for many years it is possible to superficially identify the two races with some certainty. Perhaps the history of the region should give an indication.

History of the Tanqua Karoo

The farm is situated in the Tanqua Karoo, the name probably being derived from Sanqua” indicating that the San were the first inhabitants of the region.  There is also a river bed named the Tanqua. The region is also known as the Onder Karooand Ceres Karoo. The Khoi subsequently moved into this area with their livestock and the competition for pasture and water resources became intense. This was before the days of the wind pumps and the extraction of subterranean water. The Trekboers expanded into the interior in large numbers in the 18th century. In the 1800s this was a contested area which existed between Touwsrivier in the south, to the Hantam Mountains in the north and the Roggeveld escarpment inland. It was only possible to farm small stock in this area if a farmer was able to move across the boundaries between the winter and summer rainfall area in different seasons – which still is pertinent today.  Farmers still take their sheep down to the Tanqua Karoo in winter to escape the extreme cold of the escarpment. The demarcation between winter and summer rainfall regions is not a precise line but rather a shifting corridor in which there might be little rainfall in both the winter and summer months. This all-season rainfall corridor coincides roughly with the line of the interior escarpment. Sheep and goats could survive in the winter rainfall region during winter but could not remain there when it dried out and the heat became intense.  They would be moved further inland onto the higher escarpment in summer. 

This region is subject to cycles of excessive rain followed by drought which resulted in competition for grazing amongst the Trekboers, Khoi and San hunters. This became an area of continual conflict for about a century. An example of this conflict is a report by Field Sergeant Charl Marais in September 1779. Although a comprehensive record does not exist for the climate, flora and fauna, some information is available about the rainfall in the region. Historically, the years 1700 to 1704 were years of poor rainfall but normal rainfall returned in 1705. Heavy rains in 1706 caused a great loss of livestock. July 1715 was exceptionally cold and wet and was associated also with an unknown cattle disease which led to high mortality. A severe drought in 1800 was followed by substantial rainfall in the Roggeveld in 1803 with many flowing streams. But during 1805 an extreme drought was experienced in the Tanqua Karoo. In contrast, there is a record that in one year the summer rains were so widespread that they swept through the Karoo to the sea. 

Fauna and Flora

Visiting this farm regularly throughout the years unveiled the effect of climate and weather on the ecology of the area. Nothing stays the same from year to year. An insect species which is prolific one year may disappear totally for a number of years before reappearing again when conditions are suitable. A good example is the emerald fruit beetle (Rhabdotis semipunctata) which can be found cavorting on the flowers of Vachellia karroo. Prior to the long drought they had been prolific but disappeared until V. karroo flowered again due to the heavy rains in 2022/3 [Fig.12].

Fig. 12. The emerald fruit beetle Rhabdotis semipunctata on Vachellia karroo flowers.

Botanists visiting the farm over a weekend were able to identify 98 species of plants – and many more have yet to be identified. Yet honeybees are rarely seen on flowers despite hours of hiking each day on the farm. However, they are desperate for water in the hot summer months and will visit artificial pools of water. Despite masses of vygies of various species flowering in spring, it is rare to see a single honeybee visiting them. It appears that the amount of forage in the worst month greatly influences the carrying capacity of wild colonies in an area. Honeybees have been observed visiting Haemanthus coccineus, Brunsvigia bosmaniae and various species of Asteraceae.  

Fig. 13. Succulent Tylecodon wallichii plants in flower.

Many species of succulent plants appear to have niche pollinators, many of them solitary or sub-social bee species. Tylecodon wallichii produces an exceedingly long raceme which towers above the karroid scrub. It attracts a Xylocopidae (carpenter bee) species which flies from the one exposed flower stalk to the next which is clearly visible in the landscape as they project well above the karroid scrub [Fig.13]. There are also seven species of carrion flowers (Stapeliads) whose flowers produce a stench of either sweat, urine, faeces or a rotting cadaver by which they (deceptively) attract flies without any reward which pollinate them [Fig.14]. This makes at least a subset of plants independent of bees for pollination. The pollination of these plants is further amplified by the placing of all the pollen in a sac (pollinium) which the fertilising insect transfers. A similar strategy is employed by orchids, of which only one species has been found on the farm. The Gorteria diffusa uses a different strategy to attract insects.  By expressing pigments on the petals, the spots which resemble insects, they attract the passing fly to visit the plant and it becomes a pollinator. Nevertheless, bees remain important for the pollination of some of the plants whether they flower in the summer or winter. 

Fig. 14. The carrion flower, Hoodia pilifera var. pilifera which attracts pollinating flies.

Occupation of the region by humans has undoubtedly affected the ecology of the region; especially with continual occupation and the introduction of livestock. The farm has not had stock on it for 20 years (except occasionally those of the neighbours which find an opening in the fence) but there are resident rhebok, duiker, steenbok and three pairs of klipspringers. Baboons regularly pass through and leave a tell-tale of the chewed roots of Euphorbia rhombifolia [Fig.15].

Fig. 15. Chewed remains of the roots of a Euphorbia rhombifolia plant uprooted by baboons.

Game abounded in this region before firearms appeared.  Colonel Robert Jacob Gordon in 1777 attested to the large number of lions in the Tanqua Karoo.  Augusta de Mist (1803) also commented on the large numbers of lions, leopards and hyenas and was enthralled to see a flock of about 300 ostriches on the southern plain below the Rooiberg which forms the one boundary of the farm. The Bokkeveld in fact was named after the scattered herds of springbok which migrated from the interior into this territory at times. The only accommodation on the farm is an old skeerhok(shearing shed) in which to sleep from which a beautiful view over the veld can be seen as the sun sets [Fig.16].

Fig. 16. Sundowners in front of the ‘skeerhok’.

There is no doubt that weather and climate have a large effect on honeybees which expand in strength and number of colonies in favourable years. During unfavourable times, the honeybees decrease in numbers and possibly abscond.  Judging from trap boxes placed out each year in the Swartland, drought can drastically curtail the numbers of migrating reproductive swarms. The banded bee pirate, Palarus latifrons, occurs on the farm and there are also generalist arthropod predators such as the Asilidae robber-fly [Fig.17]. However, although honey badgers have a wide distribution across Africa they have not been observed on the farm. 

Fig. 17. Asilid robber with a honeybee captured on the Tanqua farm.

The interesting observations over approximately two decades need further study before the extensive development of facilities to generate electricity by harnessing wind or sunlight overwhelm the area.  The Perdekraal Oos wind farm is located within 10 km of the farm where the observations have been made. Several other developments are envisaged in the Renewable Energy Development Zone (REDZ).  The environmental impact studies for this development and other developments under consideration in the region reflect relatively short periods of study and do not provide detailed information on the waxing and waning of the numbers of bees and their nests, the interaction of the capensis and scutellata bees as well as other related pollinators.  It is likely that periods of more summer rain have temporarily favoured scutellata bees while the capensis bees can be viewed as the predominant inhabitants of this region. It is also evident that there is an interaction between the bees and flora as well as fauna, including the provision of nests by aardvark burrows. 

Cape Point Cape bee sanctuary

Because of the high numbers of Savannah bee hives that were being brought within the natural Cape honeybee distribution area, there was a fear of a threat to the existence of the Cape bee. It was decided to found a sanctuary for the Cape honeybee where they would be protected from hybridization. A cursory survey was carried out in the proposed sanctuary of Cape Point in1980. On the recommendation that no honeybees were present in the area, a decision was made to introduce colonies of Cape honeybees. These were fortunately located within the heartof the Cape bee distribution at Fernkloof (Hermanus) and De Hoop Nature Reserve and approximately six colonies were introduced.

Yet the Cape bee has always occurred in Cape Point – one of the oldest maps of the Cape Peninsula having a location named Bynes, alludes to this. Over the past 8 years of research in locating and analysing wild colonies in Cape Point, 98 wild honeybee nests were recorded – and more still have yet to be located. Then how was the incorrect assessment of the total absence of honeybees in Cape Point arrived at? It appears that there was one visit on a specific day only and a cursory search of bees on flowers was made. At certain times of the year in Cape Point, masses of various plants can be in flower but no bees are seen visiting them. Honeybees gauge which flowers are most cost effective and will visit those despite other species in full bloom. Also, the carrying capacity of an area appears to be determined by the level of resources available in the month(s) of dearth. Thus the timing and duration of the survey becomes of utmost importance if it is not to become a mere snapshot in time!

The pressure on development of the land and the changes in climate should compel us to document our unspoilt areas in detail, to perform comprehensive EIAs and to monitor the impact of the developments for all our flora, fauna and landscape. 

Zig-zag emperor moth larvae, Gonimbrasia tyrrhea, on Vachellia karroo.

The authors at work:


Allsopp, M.H. (1992). The Capensis calamity. South African Bee Journal 64(3):52-55.

Anderson, R.H. (1980). Cape Honey-bee Sanctuaries. South African Bee Journal 52(1):3, 5-9.

Beekman, M., Allsopp, M.H., Wossler, T.C. and Oldroyd, B.P. (2008). Factors affecting the dynamics of the honeybee (Apis mellifera) hybrid zone of South Africa. Heredity 100: 13-18.

Cooke, M.J. (1992). Turnabout is fair play –Cape bee invades African bee territory. American Bee Journal 132(8):519-521.

Crewe, R.M. (1984). Differences in behaviour and morphology between capensis and adansonii. South African Bee Journal 56: 16-20.

Hepburn, H.R. (1991). Geography of the Cape honeybee based on laying worker performance. South African Bee Journal 63(3):51-59.

Hepburn, H.R (2002). Apis mellifera capensis: Concept and reality. South African Bee Journal 74(2): 37-39, 53-62.

Hepburn, H. R. and Crewe, R.M. (1990). Defining the Cape honeybee: reproductive traits of queenless workers. South African Journal of Science 86: 524-527.

Hepburn, H.R. and Crewe, R.M. (1991). Portrait of the Cape honeybee, Apis mellifera capensis. Apidologie 22: 567-580.

Hepburn, H.R., Radloff, S.E. and Fuchs, S. (1998). Population structure and the interface between Apis mellifera capensis and Apis mellifera scutellata.  Apidologie 29: 333-346.

Johannsmeier, M.F. (2005). Sweet thorn (Acacia karroo): A puzzling beeplant. South African Bee Journal 77(1): 16-18.

Martin, H. (1999). The Sheltering Desert. A.A. Donker, 324 pgs.

Moodie, G.D. (1983). Reasons for the survival of the Cape bee. South African Bee Journal 55(2): 34-37.

Moritz, R and Kauhausen, D. (1984). Hybridization between Apis mellifera capensis and adjacent races of Apis mellifera. Apidologie 15(2): 211-221.

Passmore, N.I. and Carruthers, V.C. (1979). South African frogs. Witwatersrand University Press, 270 pgs.

Penn, N. (2005). The Forgotten Frontier. Double Storey Books, Cape Town, 388 pgs.

Thomas MM, Rudall PJ, Ellis AG, Savolainen V, Glover BJ.  (2009) Development of a complex floral trait: the pollinator-attracting petal spots of the beetle daisy, Gorteria diffusa (Asteraceae).  American Journal of Botany 96(12): 2184–2196.