The results of clinical trials have suggested that ultrasound irradiation can enhance the effectiveness of thrombolytic agents and increase the recanalization rate for acute ischemic stroke patients [1]. Because of the risk of cerebral hemorrhage [2], a transcranial ultrasonic treatment device that uses a frequency $500 kHz is under development [3,4] to lower mechanical and thermal risks. Development of such a device requires knowledge of the transmissivity of ultrasound through the temporal bone, which is a thin part of the human skull ($a few millimeters thick). To clarify the transmissivity dependence on bone thickness, we experimentally measured transmissivity through a bone-phantom plate in water (which has acoustic properties similar to those of soft tissue). Although our experimental results were almost consistent with those of a model calculation based on normal incident plane waves, we discerned a definitive discrepancy between them. In this Short Note, we report on a process for pursuing the cause of this discrepancy.In the case of normal incidence of a plane wave to a bonephantom plate in water, the energy transmissivity is given by the formulawhere k b is the wave number in the bone-phantom plate, d is the thickness of the bone-phantom plate, Z w is the acoustic impedance of water, and Z b is the acoustic impedance of the bone-phantom plate (for the derivation, see, for example, [5]). The effect of absorption by the bone-phantom plate was taken into account by the substitution k b ! k b À i f , where is the absorption coefficient and f is the frequency. Figure 1 shows the transmissivity calculated by using the above formula as a function of bone-phantom plate thickness. There is a peak around 3 mm, which corresponds to half the wavelength of a 500 kHz wave in the bone-phantom plate.In our experiment, we used 20 bone-phantom plates (of sound speed = 2,884 m/s, density = 1,664 kg/m 3 , and absorption coefficient = 46.3 Np/m/MHz), ranging from 0.6 to 4.4 mm by 0.2 mm steps. Almost all Japanese temporal bones are considered to be within this range. The plates were special order products, and the acoustic parameters were measured by the manufacture. The parameters of the plate were similar to those of human skull [6]. A transducer, whose surface is a disk of 24 mm in diameter, was located 12 mm from the bonephantom plate in a water-filled tank. Ultrasound entered the bone-phantom plate almost normally. The transmitted wave was measured by a needle-type hydrophone (Onda Corporation, Sunnyvale, CA, USA) at a position 60 mm away from the transducer surface on the central axis of the disk. Figure 2(a) shows the experimental result. There is a peak around 3 mm, as expected. Equation (1) can explain the overall tendency. However, we found a dip at 3.2 mm, which is not expected by using Eq. (1).We repeated the experiment several times, and the 3.2 mm dip appeared each time. We investigated the reason for this reproducible 3.2 mm dip. At first, we suspected the quality of the material; one 3.2 mm plate might diff...