One of most important characteristics describing forest is their vertical structure (number of layers). This information is collected during periodical forest inventories. The inventory procedures are time-consuming and can provide only general information, describing relatively big parts of forests. There are still a lot of research activities on the search for methods providing more accurate information. In past aerial laser-scanning was successfully adopted by foresters to determine forest height, number of trees and biomass volume. There was suggested also to use ALS to detect forest vertical structures. This paper describes research made in an Scotch pine (Pinus silvestris) stand with some deciduous species in order to check whether it is possible to detect an forest understory with ALS data. An analysis of histograms describing vertical structure of laser-scanning clouds in half-meter layers was done. Histogram shapes were described with 7 special developed variables V1÷V7. It was found, that there is possible to detect the presence of forest understorey with one of variables.