A few years ago we were asked to investigate the allegation that three oak trees had caused a house at Ramsey Heights, Cambridgeshire to subside.
Samples of the sub-soil were collected adjacent to the house and it was found that the silty peat immediately below the foundations was highly plastic and had a high shrinkage potential.
Work at the Building Research Establishment (BRE) has shown that highly shrinkable soils in southeastern England can undergo signiยฎcant seasonal ground movements to a depth of about 500 mm, even in the absence of trees. The maximum vertical shrinkage that can be achieved by such soils in southern Britain is said to be about 50 mm . Most of the movement takes place within 1000 mm of the surface and is greatest in the late summer or autumn: however, Crilly et al. made measurements showing that dessication can extend to in excess of 3 m in the proximity of poplars. During winter the soil is recharged with moisture by rainfall and the soil usually swells to regain its original volume.
In the presence of trees the summer soil moisture deยฎcit may extend to greater depths and might not be recharged fully during the winter . Ward [1,5] concluded that on highly shrinkable clay soils, trees which transpire large amounts of water in the summer, such as oak, poplar, willow and cherry/plum, are capable of damaging buildings at distances equivalent to the height of the tree. Accordingly, BRE Digest 298 , recommends that such trees should not be closer to a building than the maximum height of the tree.
At Ramsey Heights the three oak trees nearest to the house were 8, 13, and 13 m tall and well beyond the safe distances given in the Building Research Establishment's guidelines.
No signs of oak roots were found in any of our sampling boreholes: however, a single oak rootlet was recovered by the civil engineers engaged to repair the damage caused by subsidence. The rootlet was subjected to a starch iodine test by Richardson's Botanical Identiยฎcations of Reading who showed that it was `recently alive', i.e. the test proved positive. Live roots with starch reserves typically stain blueblack in this test. Dead roots, which may still show structure but lack starch, should not stain blueblack (Ref. , p. 3). It appeared from this ยฎnding that the subsidence could have been caused by an