Chase.] 4:10 [April 16, 



LUNAR-MONTHLY RAIN- FALL IN THE UNITED STATES. 



By Pliny Earle Chase, Professor of Physics in Haverpord 



College. 



{Bead before the American Philosophical Society, April 16, 1875.) 



When the Meteorological Department of the Signal Service Bureau was 

 first organized, I believed that the extent of territory embraced by the 

 observations would soon furnish material for useful generalizations, in 

 respect to the importance of climatic influences which many regard as 

 either problematical, or wholly insignificant. 



If any considerable improvement in our present system of weather 

 forecasts should ever become possible, it will doubtless be brought about 

 by a fuller understanding of cyclical changes. Howard and Sabine long 

 ago showed that barometric pressure and magnetic force are sensibly af- 

 fected by the moon, and the cumulative eifect of undulations is such that 

 the daily atmospheric tides, though singly of small magnitude, may, by 

 regular succession, lead to such blendings of currents as will produce 

 cyclical winds and storms. By my numerous comparative investigations 

 I have shown that, while there is a great discrepancy in the forms of the 

 lunar rain curves at different stations, the discrepancy is no greater than 

 is found in the solar curves. I have also shown that there is a likeness 

 between tlie curves for difierent independent periods, at the same station, 

 which cannot be attributed to chance, such likeness being most striking, 

 and the inflections of the curves being greatest where the lunar-tidal 

 forces are strongest. 



Any normal lunar, or planetary, wave-producing influence may be 

 greatly obscured by local or accidental disturbances. The daily an- 

 nouncements of "probabilities" often seem to fail in a given locality, 

 when the weather map shows that they are wonderfully verified in an 

 entire region. So a lunar disturbance which would ordinarily bring rain> 

 may be marked by cloud or wind at some stations, while, if we had re- 

 ports from the entire disti'ict, we should find a general prevalence of rain. 

 We may, therefore, look for results from observations at a large number 

 of stations, extending over only a few jears, analogous to those which 

 would be shown in a long series of yeai's, by the observations at a single 

 station in the same district. 



The influence of the Rocky Mountains upon our storms has been well 

 known since the days of Redfleldand Espy. The intersections of normal 

 winds, near the base of those mountains, as well as the analogous inter- 

 sections which occur in the West Indian birthplace of tornadoes, I have 

 pointed out in a previous paper. In neighborhoods where there is a 

 natural tendency towards a blending of currents, cumulative tidal influ- 

 ences may be supposed to have a special efficiency. 



Influenced by these views, I have examined the morning weather maps 

 for the past three years, tabulating, in accordance with the moon's age, 

 both the number of reporting stations and the reported rain-fall upon 



