A tracer method has been used to determine the relative amounts of phosphorus that a range of short-term crops and fallows derive from different parts of their root zone when growing on a phosphorus deficient latosol. The short-term crops made little use of sub-soil phosphorus: millet derived 16% from below 12 in. after 95 days growth; and maize derived 7% and pigeon pea 11% from below 10 in. after 80 days. Pigeon pea feeds much closer to its base than the cereals until the 50th day. Since millet absorbed little phosphorus from subsoil placed in a layer only 5–10 in. deep, it is concluded that the poor uptake of subsoil phosphorus is due to low availability, rather than a natural surface feeding habit.

Pigeon pea in its second year extracted little more phosphorus from the subsoil than it did in the first. On the other hand, in a natural savanna both the perennial grasses and the associated dicots derived over 30% of their phosphorus from below the 10 in. layer. The amount corresponded with the proportion of dicotyledonous roots in the subsoil, but exceeded the corresponding proportion of grass roots. The greater proportion of subsoil phosphorus in the savanna may be due to lower availability of phosphorus in the topsoil, which in contrast to the other plots did not receive fertilizer.