The results of tapping 507 seedlings of rubber tree (Hevea brasiliensis Muell.-Arg,), 24 to 27 years old, are presented. The plants were studied in two groups, one containing 264 plants, the other 243. Based on the mean yield of the groups, respectively 31.55 and 16.79 cc of latex per tapping, it is believed that the plants of the first group were larger than those of the second. The tapping system adopted was a half-spiral cut, tapped every other day. The plants of the first group were tapped from May to October, 1944, averaging 58.25 cuts; those of the second group were tapped in September and October, 1944, averaging 26.93 cuts. The mean yield per tree was very variable, from 2.0 to 238.2 cc of latex per tapping in the first group, and from 0.8 to 107.6 cc in the second. The best plant of the first group (0.38% of the population) yielded as much as 2,86% of the total latex production, ond in the second group the best plant (0.41% of the population) yielded 2.64% of the latex. The data showed that after the first six tappings the best yielding plant was the n. 179; this result was confirmed after six months tapping. The study of the data showed the following results, respectively for the 264 and 243 plants populations. a) 10.22 and 10.29% of the plant population yielded 32.21 and 37.75% of the total latex production; b) 25.00 and 25.10% of the plant population yielded 54.92 and 61.41% of the total latex production; c) 50.00 and 50.21% of the plant population yielded 80.19 and 82.52% of the total latex production; d) 75.00 and 74.90% of the plant population yielded 93.92 and 94.31% of the total latex production. These results show the feasibility of using unselected seeds as commercial planting material, initially at very high density in the field, then eliminating the greatest majority of young plants (low yielding) in order to maintain, at production age, only the best yielders. A plantation of seedlings as those of the first studied population would yield only 632 kg of dry rubber per year, per hectare (at density of 400 plants per hectare), at the age of 24 to 27 years. Theoretically, if similar seedlings were planted at high density, the elimination of the young worst 75% of plants would increase the production to 1288 kg per hectare (at 400 plants density), and to a more drastic elimination would correspond a better average yield per plant and, consequently, a better production per unit area.