The applied biological control of Panonychus ulmi in apple orchards in Brazil began in Vacaria (RS), in 1992 by rearing Neoseiulus californicus in greenhouses and releasing them into infested areas. The balance between both mites took place week Is after the release. The aim of this work was to evaluate the effect of N. californicus density to be released in apple orchards for the control of the red mite. A commercial orchard in Fraiburgo (SC) was chosen and populations of 50,000, 100,000 and 150,000 phytoseiidae per ha were released. These populations were from the commercial breeding centre located at Renar Macas, in Fraiburgo. The damage, due to the feeding of P. ulmi was measured in terms of the range of brownish symptoms. The monitoring of hibernating eggs was carried Out in May, July and September. The treatment with 150,000 phytoseiidae reduced the Population of red mite 16 days after the release, while the other treatments required three weeks. The damage to the leaves was less in the treatment with 150,000 phytoseiidae. The monitoring of hibernating eggs using 50,000 and 100,000 phytoseiids/plot took place in May and it was 75% and 69% greater than in the plot using 150,000 phytoseiidae. It can be concluded that the release of 150,000 N. californicus was more efficient in the control of P ulmi in the reproductive phenological stage of the apple tree.