One significant feature of ultrasonic diagnostic equipment is that it displays a cross sectional image as moving images in real time. The author attempts to estimate motion distribution in biological tissues by an optical flow method from a series of B-scan images. The method of optical flow determines the motion vector from the images before and after movements under linear approximation of gradients of the gray level. However, only a low accuracy estimation is possible in the case of a long displacement or a large fluctuation in gray level distribution which cannot be from linear approximation. Thus, the motion vectors on the contour lines are calculated to increase the accuracy, because the neighboring points on the pixel of the contour better approximate linear gradients. Further, Gaussian filtering is performed to increase the accuracy. This method is applied to fetal movements, liver tissues movements, and heart movements. As a result, the distribution of the motion vectors indicates the amount of motion of fetus, liver tissues and heart to be reasonably expected. Therefore it is possible to estimate the motion of biological tissues by applying the optical flow method to the B-scan images.