side, where pressure is at a maximum, or the side from which bad weather might be expected. The most general and popular use of this instrument is that it yields records of all the great seismic disturbances in the world. From the character of a seismogram you can judge of the magnitude of the earthquake it represents. You can frequently say where it occurred, and when it occurred. This information is usually obtainable long before the arrival of cablegrams which, if they emanate from a devastated district, suffer not only delay, but may convey exaggerated and alarming impressions. Seismograms written by our earth have frequently extended, confirmed, or disproved telegrams written by man. To pressmen and the inquiring public seismograms have an increasing importance, and Africa, although it is the poorest earthquake-producing continent in the world, can with advantage to itself report upon disasters off and beyond its shores. Large earthquakes are announcements of accelerations in hypogenic activity. When this ceases, and that which is epigenic becomes paramount, it would seem that the ultimate effects of surface denudation in general would be to reduce continents to sea level and to wipe out surfaces which are habitable. The panacea for such a fear rests in the idea that as mountains are washed down to load our sea boards, these may sink, whilst the high lands, which have been lightened, would be buoyed up. By such a process isostacy would save the features of our world. HUMIDITY CHARTS OF THE UNITED STATES. BY KENNETH S. JOHNSON, HARVARD UNIVERSITY. (Received through Professor R. de C. Ward, Harvard University, and communicated by J. R. Sutton). [ABSTRACT.] Loomis in 1880* published a chart with relative humidity lines for a few stations east of the Rocky Mountains. There were but four such lines, and these were based on only one month's recordJanuary, 1875-but the purpose which Loomis had in mind was accomplished, namely, to show that on the east side of the Rocky Mountains "there is a narrow belt of territory where the mean relative humidity is less than one-half." In 1902 the U.S. Weather Bureau published † three charts of relative humidity, one for the year, one for January, and one for July. The charts are incomplete in that they do not cover all the months. Moreover, they are based on records of varying length, ranging from 4 to 14 years. The present charts for all the months and for the year are based on data for the uniform period of 14 years, from 1888-1901 ¶ In a few parts of the country, where the stations are scattered, records of slightly shorter periods have been used in determining the position of some of the lines, but in no case have the short period records been given equal weight with those of the 14-year period. The curves are drawn for differences of 10 per cent. only; for the sake of clearness in presenting the main facts. Furthermore, the length of the records is not sufficient to warrant greater detail. Among the most striking features shown on the charts is the uniform high relative humidity along the coasts, in contrast with which is the extremely low relative humidity in the S. W. interior, especially in Arizona, Nevada, S. E. California, and the adjacent districts. While the high relative humidity on the coast remains fairly constant throughout the year, the low in the interior basin (or, as it used to be called, the Great American Desert) becomes more marked as summer approaches, thereby increasing the already strong relative humidity-gradient between the Pacific coast and the interior. Another interesting fact is the annual movement of the lines in the northern part of the interior basin. These travel north in summer, reaching the northern limit in June or July, returning south or perhaps entirely disappearing in winter. Among other features worthy of note is the distorting effect of the Great Lakes, where the relative humidity tends to remain at or near 80%. an effect quite similar to that produced by the oceans. Also a thing which rarely occurs-there is a prevailing north and south trend of the lines over the Great Plains in the vicinity of the 100th meridian, showing that the relative humidity there does not vary with the latitude. * American Journal of Science, Third Series, Vol. XX., p. 22. |