For a study of population trends of Diarthrothrips coffeae, Williams, on coffee in Kenya, and to assess the value of sprays for its control, a method of sampling the population was required. This insect feeds on the leaves, green shoots, and berries ; but by far the greater part of the population lives on the leaves. Attempts to count the insects on the leaves in the field, by naked eye or with a hand lens, proved difficult and inaccurate. Two other methods were tried, those of E. A. McGregor (Pan-Pacif. Ent., 3, 1926, p. 29) and J. W. Evans (Bull. Ent. Res., 24, 1933, p. 349).The former proved unsatisfactory for several reasons such as the failure of the insects to forsake the leaves, the collection of debris on the grease tending to conceal insects, and the difficulty of counting the insects in the grease. The method of Evans was designed originally for obtaining the thrips from blossoms, but an examination of the method was made with the coffee thrips. Fairly consistent results were obtained with it, but the figures were low and it was clear that the sampling was not complete. It was found that a number of insects remained among the leaves and failed to reach the paper at the bottom of the cylinder. There are other disadvantages ; for example a large number of the cylinders are necessary to carry out a single spraying experiment, they are usually difficult to come by and the work of counting has to be done in the field, requiring a rather elaborate provision of tables and microscopes. Further it was found that even a light breeze was apt to cause loss of thrips during the counting.