Abstract
The optimal root system architecture for increased tree anchorage has not yet been determined and in particular, the role of the tap root remains elusive. In Maritime pine (Pinus pinaster Ait.), tap roots may play an important role in anchoring young trees, but in adult trees, their growth is often impeded by the presence of a hard pan layer in the soil and the tap root becomes a minor component of tree anchorage. To understand better the role of the tap root in young trees, we grew cuttings (no tap root present) and seedlings where the tap root had (-) or had not (+) been pruned, in the field for 7 years. The force (F) necessary to deflect the stem sideways was then measured and divided by stem cross-sectional area (CSA), giving a parameter analogous to stress during bending. Root systems were extracted and root architecture and wood mechanical properties (density and longitudinal modulus of elasticity, E L ) determined. In seedlings (-) tap roots, new roots had regenerated where the tap root had been pruned, whereas in cuttings, one or two lateral roots had grown downwards and acted as tap roots. Cuttings had significantly less lateral roots than the other treatments, but those near the soil surface were 14% and 23% thicker than plants (+) and (-) tap roots, respectively. Cuttings were smaller than seedlings, but were not relatively less resistant to stem deflection, probably because the thicker lateral roots compensated for their lower number. Apart from stem volume which was greater in trees (+) tap roots, no significant differences with regard to size or any root system variable were found in plants (-) or (+) tap roots. In all treatments, lateral roots were structurally reinforced through extra growth along the direction of the prevailing wind, which also improved tap root anchorage. Predictors of log F/CSA differed depending on treatment: in trees (-) tap roots, a combination of the predictors stem taper and %volume allocated to deep roots was highly regressed with log F/CSA (R 2 = 0.83), unlike plants (+) tap roots where the combined predictors of lateral root number and root depth were best regressed with log F/CSA (R 2 = 0.80). In cuttings, no clear relationships between log F/CSA and any parameter could be found. Wood density and E L did not differ between roots, but did diminish with increasing distance from the stem in lateral roots. E L was significantly lower in lateral roots from cuttings. Results showed that nursery techniques influence plant development but that the architectural pattern of Maritime pine root systems is stable, developing a sinker root system even when grown from cuttings. Anchorage is affected but the consequences for the long-term are still not known. Numerical modelling may be the only viable method to investigate the function that each root plays in adult tree anchorage