Abstract
1) The present paper dealt with the problems of line transects in animal ecology. Several methods of the use of line transects were compared with one another and criticised both from the theoretical and practical point of view.
2) The densities of certain woodland bird populations were estimated by census and sample counts at the same time and in the same place. When the results were compared a serious discrepancy was observed between the density determined by complete census and the density determined by sample counts and mathematical calculation.
3) It was concluded that the discrepancy was probably due to the variability of conspicuousness of the birds. The influence of conspicuousness of the animals on the results obtained by sample counts was discussed under the assumption that conspicuousness may vary not only with the species, with the observer, with the habitat, and with the distance, but also with the speed of the animals.
4) The methods of line transects were classified from a practical stand point according to what kinds of animals are to be sampled. The precautions to be taken in the application of each method was also suggested.
