Africa. In the island of Bali, near Java, a woman who is so unfortunate as to bear twins is obliged, along with her husband, to live for a month at the sea-shore or among the tombs, until she is purified. The Khasias of Hindostan consider that to have twins assimilates the mother to the lower animals, and one of them is frequently put to death. An exactly similar belief prevails among some of the native tribes of Vancouver Island, Among the Ainos, one of the twins is always killed, and in Arebo, in Guinea, both the twins and the mother are put to death.
Father Secchi on Solar Spots.—In summing up the results of his observations on solar prominences and spots from April, 1871, the Roman astronomer, Father Secchi, states that since that date there has been a very marked diminution both in the number of groups of sun-spots and in their area—a result of the eleven-year period, the maximum having occurred about 1871. But he has observed the same diminution in the number of prominences: in 1871 the daily average was about fifteen, while now it is six or seven. In the same period the number of groups of sun-spots in each rotation has decreased from about twenty-five to eight, and the mean area has diminished to about one-fifth. Further, the prominences are now very rare near the poles. Secchi further remarks on the discordance between his results of 1852, showing a difference of temperature between the solar equator and poles, and those of Prof. Langley, and infers that there has been a change in the sun in this respect, consequent on the decrease of solar activity. He objects to Langley's method of moving his thermopile to different parts of the image instead of moving the telescope so as to bring the points of the image in succession on the thermopile, and thus to avoid differences of inclination to the axis of the lenses.
Prof. Loomis on the Storms of the United States.—This eminent meteorologist presents, in the July number of the American Journal of Science, his third paper on storms, founded on the weather-maps of the Signal-Service. He is now able to confirm what was stated in his previous papers in regard to the general progress, direction, and barometric phenomena of storms in the United States. These papers of Prof Loomis are admirable in method, and of very great value. The general direction of the storms which traverse the United States is found to be a little north of east, but varies somewhat with the seasons. Thus, July storms are most southerly in their direction, being a little south of east, those of February being most northerly. Rarely, storms move for a time northward or southward.
By direction of a storm is meant the movement over the country of the whole storm, not the direction of the winds, and its progress varies greatly in rapidity. The average velocity during the past three years has been 26 miles per hour, the storms of August being slowest, those of February and March being most rapid. The storm of February 22, 1874, moved at the tremendous rate of 53.3 miles an hour, or 1,280 miles in a day.
It is also shown that the progress of storms is not uniform throughout the day, but has a uniform daily variation. The velocity is greater by 25 per cent, from 4.35 p. m. to 11 p. m. than during other portions of the day, and this is constant during each month of the year. The greatest velocity occurs at about 7 p. m. without apparent relation to the wind's velocity, or absolute temperature. "But," the professor observes, "it is the time when the temperature of the day is declining most rapidly." Now, this change of temperature has direct relation to those conditions which cause precipitation and extend the rain-area. By reference to a former paper of Prof. Loomis it will be seen that condensation in front of the storm-centre is one means by which a storm progresses. It is continually making up in its front where the air is vapor-laden, not in its rear where the air has been deprived of its vapor.
It will hardly admit of question that the velocity of a storm's forward motion is usually accompanied by an extension of the rain area in the direction in which the storm progresses. The average extent of this area in front of the storm-centre during three years is found to be 542 miles. Now, if this be increased 100 miles, the velocity