Atmospheric temperature has been continuously recorded by platinum resistance thermometers at the heights of 1.3m and 16m above ground level in the Mizusawa Geodetic Observatory since June 1973 with the aim of improvement of temperature correction for the optical distance measurement. Remarkable nocturnal inversions are observed especially on fine days. At about 6h-7h in the morning and 16h-18h in the afternoon, the temperature difference between the upper and lower position becomes zero. The temperature difference remains nearly zero for many hours when it is cloudy. Such a condition might be preferable for the temperature correction of the optical distance measurement.
According to the theory of plate tectonics, the earth's crust is strained by the motion of upper-mantle lithosphere. A large earthquake is thought to be a strain release . A belt-conveyer model introduced by Utsu  well explains the earthquake phenomena occurred in relation to the motion of lithospheric plate. Assuming that the timely variation in horizontal strain is expressed by a Markovprocess, the associated equations are derived. The results of Mogi's rock fracture experiments  are applied to the hazard rate of earthquake occurrence .
Effects of oceanic tides on earth tides at the same stations reported in previous parts have been re-estimated using the Green's functions by Farrell (1972). The present results for the Gutenberg-Bullen A earth model agree fairly well with the previous results estimated by the Kuo's theory and Longman's Green's functions. Existence of remarkable discrepancies between observed and calculated earth tides in several cases indicates that it is important to investigate the combined response of each instrument and observation site to earth tides not only for the study of earth tides themselves but also interpretation of crustal deformations observed by the instrument.
At Mikawa Crustal Movement Observatory, continuous observations of crustal deformation have been made since January, 1973. The instruments used in the observatory are three silicatube extensometers of horizontal component and a set of watertube tiltmeters specially installed in the gallery. Differential transformers are used for all the instruments. Each silicatube extensometer has two or three detectors so that it is possible to discuss the irregularity of strain. The observation of strain in the period January, 1973 to December, 1973, seems to indicate the following features: the def omation is uniform for short period but not uniform for long period. Continuous ground tilts have been observed with both watertube tiltmeters of float-type and traditional micro-screw-type installed on the same tables which are made of granite. From the observation of ground tilts with watertube tiltmeter of float-type, it is found that this instrument has enough stability for continuous observation of ground tilts.
Since April, 1973, continuous observation of ground tilts has been carried out with newly designed water tube tiltmeters of float-type at Mikawa Crustal Movement Observatory. In the present paper, details of the instruments are described. The instrument is of low cost and high sensitivity. The results obtained with this instrument indicate the high stability enough for the continuous observation of the ground tilts . Instrumental drifts are detected by traditional water tube tiltmeters of micro-screw-type installed on the same tables which are made of granite. The rate of drift is constant for a long period of time so that secular ground tilts are easily obtained .
It is important to disclose the characteristics of gravimeter in the case of precise gravity surveys. The authors have made an experiment in order to, obtain the relation between airtemperature and gravimeter drift, that is essentially the main subject to be solved in investigating the characteristics of gravimeter. As a result, the authors have concluded as follows; 1. In the case that air-temperature increases rapidly up to about 40°C from 5°C, the reading values of gravimeter decrease about 0.02 mgal. In contrast with the above case, if air-temperature decreases, the reading values have a tendency to increase. It seems likely that the gravimetric response to air-temperature change is fairly rapid, especially in the case of increasing air-temperature. 2. If a gravimeter is set in high air-temperature, the drift-rate is relatively small in comparison with the case of low air-temperature.
Land magnetic observation for more than 24 hours on each 35 repeat station in Japan were carried out in 1969 and 1970 by Hydrographic Office of Japan. Geomagnetic inclination and declination were observed by earth-inductor type magnetometer for about every an hour. Total intensity was measured every minute by proton magnetometer. Observed data were compared with those of 4 geomagnetic permanent observatories in Japan. Mainly we examined the changes of total intensity on each repeat station. Daily variation of total intensity in Japan changes considerably day-to-day and placeto-place. Those are considered to be caused by changes of equivalent Sq current vortex and anomalous variation due to local conductivity anomaly of the under-ground structure. Among them we found phase delay of an hour for durnal variations of total intensity F on the stations along the coast of Japan sea from Hokkaido to Kyushu area compared to those of Pacific side of same area. These phase delay may be caused by conductivity anomaly mentioned above. Perkinson vectors obtained from observed total intensity data of short period variations on the repeat stations are shown here. At that case, we assumed that short period variations of horizontal component H and declination D were similar to those of nearest permanent observatories.