MATERIALS AND METHODS (cont).

Irrigation was applied frequently (twice daily if needed) in the early stages to prevent moisture stress. As summer 1990 was warm and dry, watering was continued until 13 September after which no further irrigation water was supplied.

Fertiliser was supplied on five occasions: 9 May, 7 June, 23 July, 16 August and 11 September. The first three applications consisted of a 12:0:9 compound fertiliser at a rate of 20 g/m2 and the last two of ammonium nitrate at 15 g/m2 applied in solution to give as even a coverage as possible.

The trial was top dressed with the same sand as was used in construction as follows: 22 June, sown plots 2 kg/m2, turfed plots 1 kg/m2 (the higher rate on the sown plots being used to bring up the levels to the same as those on the turfed plots); 2 July, sown plots 1 kg/m2; 23 July, 6 September, 24 September and 24 October, all plots received 1 kg/m2.

Wear treatments were carried out using a differential slip wear machine fitted with golfspikes at a rate of 4 passes of the machine weekly, between 6 September and 27 November, a total of 52 passes being applied.

Data collection.
Data were collected on two occasions: before the start of wear (28-31 August 1990); and after three months of wear (3-6 December 1990).

Data collected included : ground cover and species composition; water infiltration rate; green 'speed' and hardness.

Ground cover and species composition were measured using an optical point quadrat described by Laycock & Canaway (1980); 100 points per plot were used in 20 frames of 5 points.

Water infiltration rate was measured using ponded, double ring infiltrometers, 300 mm and 500 mm in diameter for the inner and outer rings respectively. Steady state infiltration rates were standardised to a temperature of 10°C. Three measurements were made on each plot.

Green 'speed', or more correctly distance rolled, by a golf ball was measured using an inclined ramp. The ball was released from a vertical height of 200 mm with the inclination of the ramp being of 20°. The distance rolled by the ball from the end of the ramp to its final resting point was measured. A total of six roll tests were carried out on each plot, three in opposing directions.

Hardness was measured using a Clegg Impact Soil Tester (Clegg 1976). The peak deceleration of a 0.5 kg, 50 mm diameter hammer, dropped from a height of 300 mm, was recorded on impact with the turf. Five measurements were made per plot.

Data were analysed using analysis of variance and the least significant difference (LSD) calculated for p = 0.05 after testing the difference between any pair of treatment means, where analysis of variance was not possible the standard errors of the means were calculated.