Revisiting Theron's hypothesis on the origin of fairy circles after four decades: Euphorbias are not the cause.

backgroundThe Euphorbia hypothesis on the origin of fairy circles (FCs) in Namibia dates back to 1979. It proposes that the remains of decaying shrubs would induce an allelopathic interaction with the grasses and thereby cause bare-soil FCs. Here, we investigated this hypothesis based on revisiting marked Euphorbias after four decades, comparing the typical size distribution of dead Euphorbia damarana and FCs, and analyzing the spatial patterns of Euphorbias and FCs within the same drone-mapped study plots in three regions of Namibia.

resultsWe found four dead Euphorbias in the southern Giribes that were marked by G.K. Theron about 40 years ago. Those locations did not develop into FCs over this time span. However, for the four dead Euphorbias, we provide photographic evidence that grass tufts were growing at the metal pins of those decaying shrubs, agreeing with previous research findings that the soil taken from beneath dead E. damarana shrubs was stimulating rather than inhibiting the growth of grasses. In the Giribes, there were very large FCs that ranged in diameter from 13.0 to 19.1 m. By contrast, the measured dead E. damarana, including the largest plants that we could find, ranged in size only between 4.2 and 11.7 m. At Brandberg, we found particularly small FCs with diameters between 2.4 and 2.7 m but the dead E. damarana, including the smallest dead shrubs in the area, ranged in size between 4.1 and 7.2 m. Hence given these size mismatches, the decaying Euphorbias cannot induce such observed FCs in the two regions. Spatial patterns of E. damarana and FCs in the two regions Giribes and Brandberg, as well as of E. gummifera and FCs near Garub, showed a strong mismatch within the same habitat: in four out of five plots the patterns differed significantly. FCs were regularly distributed while Euphorbias were predominantly clustered.

conclusionsWe reject the Euphorbia hypothesis based on the fact that grass growth was not prevented under decaying shrubs, the size of dead Euphorbias cannot explain the size of observed FCs and the spatial distribution of Euphorbias cannot cause the specific pattern signature of FCs.