Harvesting: eight months after starting the treatments all plants were harvested, total height was measured, number of leaves was counted and leaves, stems and roots were dried separately in an oven at 70 °c for 96 hours; then the parts were weighed to obtain dry weights.

Chioropyll: before harvesting, leaf samples from eight plants per treatment were sampled to extract chlorophyll witout xnaceration, in dimethyl sulf oxide (DMSO) using the method of Hiscox and Israelstan (1979). The procedure was as follows. Five leaf disks, from each leaf (total weight ca. 125 ing) were rinsed in distilled water and iminediatelly placed in 10 nil DMSO in scintillation bottles and heated to 65 °C for 45 minutes in water bath. The extract were cooled at room temperature, and the concentration of chlorophyll in the DM50 was determined by measuring absorption at 6.45 and 663 nanometer of wave-length in a Varian Cary 219 Spectrophotometer. The Arnon (1949) equation (Spence, 1967) was used to convert the the absorption values to mg per liter of solution. Data were analysed with Statview II for one factor analysis of variance for repeated measures.

Results., and discussion

After the second month differences could be seen in favor of both treatments including ROOTS® with low and high fertilizer level. However, from the fourth month the plants receiving high fertilization plus ROOTS® started to show some symptoms of leaf burning and a stopping of height growth and leaf expansion. At this time the best groups of plants were the high fertilized control and the low fertilized plus ROOTS®. The control without fertilization showed very little growth and appeared chlorotic.

Leaf production of coffee occurs through the year, but the rate of growth varies with changes in climatic factors (Barros and Maestri, 1972). Generally, leaf growth occurs most rapidly during the wet season, with larger leaves produced than during the dry season (Cannell, 1971). Maestri and Barros (1977) reported that the rate of leaf, expansion, as well as final size of leaves follow a similar trend. These authors reported a range of mean growth rates going from 0.9 cm2 /week for the driest season to 9.2 cm2 /week for the rainy season. Because of the experimental conditions in which the experiment was carried out, it is assumed that the rates of growth and size of leaves in this study should be comparable to those reported for the wet season.

Leaf areas of the biggest leaf at two months are presented in Figure 1. So far the differences were established between the treated and the non-treated plants;