Part I is the report on the oceanographical observations of the Nakami in the summer of 1933. Though the distinguished difference of salinity was observed in the horizontal distribution of the upper layers at the oceanographical investigation of 1929 on board of R. M. S. Syumpu-maru, in the present case the salinity is averaged horizontally showing the value about 24‰. It indicates the fact that the invasion of the main sea water into the bay is obstructed. The quantities in silicate and phosphate in this summer are as much as two or three times comparing with that in 1929. Oxygen content dissolved in water is above 5c.c./L in the layers of 0 and 3m. depth and below 2c.c./L in the bottom layer.
The relation between the diatom and the nutrient salts is described in the present paper. Chaetoceras is the most predominant genus in diatom in the water of the Nakami. The quantities of nutrient salts consumed by 10, 000 of Chaetoceras per litre are calculated as following: 1.17mg/L in SiO2 and 0.072mg. in P2O5, and the ratio is 100:6.2, when the amout of solar radiation is 1, 000 gr. cal./cm2. in the open air. It is remarkable from the view-point of limiting factor that the content of silicate in the water of upper layer on 23, Sept. is decreased to zero due to the consumption of diatom while phosphate is increased more or less. On 16, Sept. there was found a vast crowd of resting spores of Chaetoceras (about 14, 000, 000/L.) in the bottom layer of St. 9 where the amount of dissolved O2 was 0.43 c. c./L. It is supposed that one of the main factors to make damage for shell-fish in “Akasiwo” is the scarcity of dissolved O2 caused by the putrefaction of dead diatom.
In connection with the recent researches of Mr. K. Takahasi* and Mr. S. Watanabe* the now already classical works of Mr. T. Kameda and Prof T. Terada on the apparent periodicity observed in statistical series have again been taken in argument and extended so as to include the not necessarily independent series, whereby, however, we have been obliged to set a definite assumption as to the distribution function, namely to be that of Gauss' normal law. When very strong correlations between successive events are assumed, the, “apparent” periodicity will pass into the “real” periodicity, under which we understand, not the fixed, usual periodicity of definite phase and amplitude, but the dynamical periodicity of the equation of motion, which can possibly be supposed to govern the phenomenon considered. The general discussion of the minimum-minimum curve of Mr. Takahasi is accompanied by serious mathematical difficulties and complications and touched only in a case, which allows the simple mathematical treatment.
In this paper we tried a systematical investigation of the correlation between the air temperature and the sea surface temperature, using the data collected by Messrs. Utioke and Yosimura, who themselves investigated the same subject and their results and conclusions are seen in their paper published in this Journal several years ago.(1) As they did, we also used the monthly mean temperature. The methods of investigation adopted by us was as follows; now we consider the anomary of the air temperature and water temperature at the sea surface for a month, and these two quantities for the preceding month, then we calculated systematically five correlation coefficients between these four quantities. In §3 we calculated these correlation coefficients, using all the data of the whole period (1911-1928 or 1913-1928), and results of the calculation are discussed in §4. In §6-§8 we studied on the annual variations of these correlation coefficients, and Fig. 9_??_15 show the curves of the variations of the averaged coefficients for four stations (Miyako-Isinomaki, Todosaki, Nosyappmisaki Erimomisaki). All results and conclusions are presented in the last section §10.