The purpose of the present paper is to analyze the distribution of sexes in the nest of Osmia excavata Alfken. For this purpose, some experiments and observations were conducted at the campus of Kyushu University, Fukuoka, in 1954 and 1955. The outlines and results of the experiments are summarized as follows: 1. In the spring of 1954, 50 females and 150 males of Osmia excavata were liberated in the field where 200 pieces of reed (15~25 cm in length and 4.5~8 mm in diameter) were set for nesting. As the results of the nesting activities of these bees, it was found that 162 out of 200 reeds provided were utilized as nests when the flying season was over. These 162 nests were composed of 73 incomplete (unhealthy) nests including those which were more or less damaged by the parasites, especially by the mite, and 89 complete (healthy) nests. The distributions of sexes in these 89 nests were shown in Table 1. 2. From Table 1, it is clear that there are four types of th e distribution of sexes, namely, (1) the first type that the nest is occupied by the males only, (2) the second type that the nest is occupied by the females only, (3) the third type that the nest is provided with a group of females basally (base means the bottom of the nest) and a group of males apically (apex means the entrance of the nest), and (4) the fourth type that the nest is provided with both females and males irregularly. Among these types of the distribution of sexes, the third type appears most abundant in number (about 64%). Therefore, it is probable to consider that this indicates an important fact concerning the distribution of sexes. 3. As a rule, a female bee of Osmia excavata completes several nests in all her life. Therefore, the marked female bees were liberated to ascertain the nesting order as well as the distribution of sexes in an individual nest. The results were shown in Tables 2 and 3. 4. If a severe disturbance, for example, such as a closing of the entrance of the nest, was applied to a nest in progress, the mother bee gives up the continuation of the nest building. After a short excitement, she begins, usually within one hour, to nest another ones successively. Therefore, the further experiments were made to analyze the distribution of sexes in these nests built after such disturbances. The results were shown in Table 4. 5. In Tables 2 to 4, it is indicated that a group of females is deposited more basally than a group of males in any nests with a very few exceptions. Then, it is proved that the mother bee produces her descendants in this manner ...... a group of females at first, followed by a group of males at second, a group of females at third, followed by a group of males at fourth, ...... 6. The author stated in his former paper that "Befo r e starting her nesting activity, the female bee copulates once. No copulation has been observed whenever the female bee begins to nest." It is believed by the further observations that a female bee copulates only once in her life. According to the author's experiments (unpublished), furthermore, the males of this species are produced by the unfertilized eggs. On the bases of these facts, it is concluded that (1) the sex of the descendant is controlled by the mother bee under a rule of the distribution of sexes, that is, a group of females is deposited basally and a group of males apically in one nest, (2) the nests of first type (all males) are derived from the virgin females or sometimes appeared about the end of nesting activities, and (3) the nests of both second (all females) and fourth (irregular distribution) types belong to the exceptions.