In a study covering the period 1949-1958 the sex ratio of births occurring in hospitals and health centers was computed in order to compare the sex ratio of the regions of India. On the basis of these recorded births the differences between regions was minimal averaging 106 live males to 100 live female births - lower in the North Zone (104) and higher in the Central Zone (108). Small sample size was thought to have resulted in these differences. Several factors affecting sex ratio at birth were explored including stillbirths parity and socio-economic factors. It was concluded that the influence of any particular factor was small and that some of the variables have a counterbalancing effect. Hospital deliveries were considered a valid indicator of the order of magnitude of the sex ratio at birth for the population at large.

Defining the studentized smallest chi-square v with parameters t, m, k, as the ratio of m times the smallest of k independent chi-square variates with t (degrees of freedom), each to t times another independent chi-square variate with m d.f., the lower 5% points of v are tabulated for different values of t, m, and k. Some possible applications of v are also indicated.

In this paper we have solved certain distribution problems connected with the simultaneous analysis of variance test [1] and have proved that the power function of this test has the monotonicity property.