Axis splitting is a widespread phenomenon in desert shrubs, and has been reported for shrubs from several plant families, both in old- and new-world deserts. It is so common in dwarf shrubs of arid environments as to be a defining characteristic of this growth form. Although anatomists described this phenomenon several decades ago, there has been only one ecological study of one species, Ambrosia dumosa. The anatomical nature of the various splitting mechanisms that have been found suggests axis splitting to be an extreme form of hydraulic segmentation. The adaptive advantage of clonal splitting in desert shrubs has yet to be determined, but it appears to be largely a risk-spreading mechanism that enables independent mortality of integrated hydraulic units (IHUs) or ramets. This should be especially advantageous in heterogeneous, water-limited environments, where soil water occurs in pockets too small to support a large shrub-genet. Clonal splitting may cause an increase in intraclonal competition among ramets, but there are also indications that at least some species possess mechanisms to reduce competition by minimizing root system overlap among ramets. Many desert shrub species that undergo clonal splitting maintain a dense clump growth form, possibly because such a growth form has positive effects on water and nutrient status of the soil and long-term effects on other soil properties.