For three years some experiments have been made to test this question. Three years ago some young heads of red clover were tied up in cloth sacks before any flowers had opened. A similar number of the same age were marked. The \ exact numbers were lost, but those heads which were kept covered produced about two-thirds as many seeds as those not covered. Very small insects may have gone into the flowers, but no bees could reach them. In 1S78, eight heads of red clover of the same age were marked. Four of them were covered with muslin bags, till all were dead ripe; the rest were left uncovered. The heads covered yielded respectively, 18, 30, 38 and 41 seeds; those uncovered yielded 46, 54, 43, 57. The total seeds from covered heads, 127; from uncovered, 200. In a similar manner, during 1879, 31 young heads of the first crop of red clover were covered before any flowers opened. Only one produced any seeds and that bore seven. At the same time 31 other young heads were marked for comparison, and yielded respectively the following number of seeds: 11, 12, 2, 35, 40, 21, 3, 1, 22, 15, 15, 20, 16, 4, 29, 3, 25, 8, 15, 22, 24, 28, 41, 10, 22, 13, 18, 18, 10, 12, 10, an average of nearly 17 seeds per head.

On August 9th, 9 heads were covered before any flowers had opened. When ripe they yielded seeds as follows, 36, 51, 45, 59, 15, 26, 47, 31, 28; an average of about 37 seeds to the head. Eleven heads which were covered yielded no seeds. The experiments of each year were made by three different trusty students. In nearly every case, the heads which were covered soon fell to the ground where they remained till ripe. This may have interfered more or less with the production of seeds.