Windthrow is a major influence on silvicultural practice in the UK. Profitability is reduced by the need to fell plantations prematurely. In recent years there has been much debate on the role of spacing and wide- spacing, or respacing has been advocated as a means of improving stability. This debate has lacked quantitative evidence. The purpose of this study was to assess the factors involved in tree stability and to use the results to. investigate the influence of spacing and stability. Using the approach of Petty and Worrell (1981) stability was first considered as a static model. The factors involved are the within canopy wind profile, the stem and crown weight distribution of the trees, the deflection curves assumed by the tree in response to wind loading and the maximum turning moment causing stem breakage or uprooting. Investigation of the influence of spacing on the resistance of Sitka spruce trees to stem breakage and uprooting showed that the maximum turning moment was positively related to the stem weight, dbh3 and dbh/wt. There was no evidence to suggest that the relationship between the maximum turning moment and the independent variables differed between spacings. Analysis of data from Forestry Commission tree pulling experiments revealed that on sites where stem breakage and uprooting occurred the turning moments causing stem breakage and uprooting were similar. The deflection curves of five Sitka spruce trees, growing on a brown earth site, were measured and compared with the theoretical deflection curves devised by Petty and Worrell (1981). The theoretical curves gave reasonable approximations to the actual curves for small deflections. Most of the experimental work was undertaken at the Forestry Commission's experimental area in Moffat Forest where equipment was installed for the monitoring of airflow above and within the forest canopy and of the tree response. Using Forestry Commission data the within canopy wind profile was obtained. Ten Sitka spruce trees were loaded until failure and the stem deflection curves, stem and crown weight distributions and the turning moments causing soil failure and uprooting were measured. The theoretical stem bending curves gave good approximations to the actual curves until soil breakage occurred. Following soil breakage two samples were released and repulled and, although the maximum resistance and the deflection at this point were unaltered, the initial force per unit deflection was halved. The critical wind speeds to cause uprooting were calculated using measured wind profiles and assuming static loading of the crown. Values obtained greatly exceeded the wind speeds recorded during a gale which caused damage. The damping ratios of the 3 trees bearing accelerometers were measured with and without crown contact. For the 2 dominants canopy contact had less effect on the damping ratio than the sway direction whereas for the subdominant the canopy treatment had the greater effect. Recalculation of the critical wind speeds of the ten tree pulling samples, using the damping ratios estimated for the accelerometer trees, and assuming a resonant response reduced the values to within the range of measured gusts recorded during a gale. Displacements of the 3 accelerometer trees, estimated from accelerometer data, were compared with simultaneous wind speed recordings. Perfect resonance was not observed but very large oscillations were noted to build up over a period of 2-3 cycles. The larger gusts (?10m m s-1) tended to produce a static response and to be responsible for the initial deflection whilst subsequent smaller gusts tended to cause the dynamic responses. In order to predict the critical wind speeds causing damage it is necessary to take this dynamic interaction into account. Theoretical stability calculations for unthinned plantations at spacings of 2.0m, 2.4m and 3.0m showed that for static loading the increase in the resistance to uprooting or breakage as a result of increasing mean tree size outweighed the greater drag force on the crown. The wide spaced stand retained its stability advantage even for resonant wind speeds. However, small changes in the within canopy wind profile and the resistance of the trees to uprooting/breakage removed the advantage gained by the wider spacing. More information is required on the influence of spacing on the dynamic response of trees to wind loading before increasing the spacing can be said to improve stability on wet soils.