December 2, 1892.] 



SCIENCE. 



31' 



The conclusion that the full moon is generally followed by a 

 decrease in rainfall is not as plainly marked as the above, but the 

 following are the total amounts in inches for the two days before 

 and the two days after full moon: — 



The third and fourth conclusions, that the wettest period in the 

 lunar month is near and before full moon, and that the driest 

 period is near and before first quarter, are distinctly marked in 

 the several mean curves. The mean result for the ten years is 

 that 6.1 per cent of the rainfall occurred on the day of the first 

 quarter and the two days before, while 13.3 per cent occurred 

 on the day of the full moon and the two days before. In inches 

 of rainfall the results for these two periods for several groups of 

 years are as follows : — 



1885, there is a short paper, in which it is shown that over this 

 country as a whole there is a preponderance of thunder-storms 

 during the new moon. While in New Haven, Conn., special re- 

 search on this question showed that in that place there was, from 

 1873 to 1880, nearly a half more rain just before and after new 

 moon than full moon. A farther investigation for this whole 

 country, also for 100 years at London, England, gave a negative 

 result; that is, no effect from the phases of the moon. In 1889 

 an investigation on the lower California coast gave a preponder- 

 ance of rain during full moon. 



It has occurred to me that it would be advisable to calculate 

 the data at Philadelphia, Pa., which is not far from Bethlehem, 

 for this question. I first computed the data for fifteen years, 

 1871-1885. and afterward for the ten years 1883-1891, with the 

 result given in the accompanying table : — 



1881-1890 

 Odd years 

 Even years 

 1881-1885 

 18h'6-1890 



Day of Full Moon 

 and two days before. 



60.80 

 33.64 

 37.16 

 20.76 

 40.04 



Day of Firat Quarter 

 and two days before. 



31.78 

 13.76 

 18.03 

 12.65 

 19.13 



The distribution of rainfall at and around the time of the 

 changes of the moon has been the element most generally studied 

 in connection with this question. In order that the records now 

 under review may be compared with others, the following are 

 given for periods of one day, three days, and five days respec- 

 tively. These are for the ten years 1881-1890 and in inches of 

 rainfall. 



New Moon 

 First Quarter 

 Full Moon 

 Last Quarter 



These figures, like those previously given, indicate that the 

 maximum rainfall occurs near full moon, and the minimum near 

 the first quarter. It Is impossible indeed to avoid the conclusion 

 that at Bethlehem, Pa., during the years 1881-1890, the distribu- 

 tion of the mean rainfall seems to have been arranged with respect 

 to the changes of the moon. If the moon really influences the 

 weather it is to be expected that a connection will also he observed 

 in other records, but it cannot be expected that the maximum 

 and minimum rainfall in the lunar month will besimilarly situated 

 in all cases with respect to the times of change. I venture fur- 

 ther the suggestion that, if the moon affects the rainfall, the 

 greatest influence will probably be found in connection with 

 thunder storms and local showers. 



It will be seen that in the first period of fifteen years there is a 

 preponderance of rain at the time of new moon, which corrob- 

 orates the result previously obtained at New Haven. In the sec- 

 ond period, for the three days about each phase the result is 

 similar to that of Professor Merriman, though the difference of 

 two inches between new and full moon is very slight. When we 

 take the five days about each phase, however, we see that the new 

 moon has 13.5 inches more rain than the full. I do not advance 

 these figures as proving any influence whatever. It must be 

 almost inappreciable if there is any at all. 



A word may be added regarding the influence of the moon in 

 driving away clouds. I have detected this apparent influence 

 many times by closely watching the moon. Of cotu'se, if this is 

 a fact, it would show that there must be a tendency to less rain at 

 the time of full moon. It should be borne in mind, however, that 

 the minimum of cloudiness occurs in the evening or before mid- 

 night, and this complicates the phenomenon. 



II. — BY H. A. HAZEN,! WASHINGTON, D.C. 



There is hardly an idea regarding the weather so firmly rooted 

 and so widespread as this, that the moon has a rather marked 

 effect in bringing about its changes. This paper by Professor 

 Merriman is a very interesting contribution to the subject. I de- 

 sire to add a little to what he says, as his conclusions are not the 

 same as those reached by myself. This matter has been thor- 

 oughly investigated in England and Europe with a negative re- 

 sult, except that there seems to be a slight influence of the moon, 

 or perhaps the tide, on the occurrence of thunder-storms, and that 

 the full moon seems to have power to drive away clouds. All the 

 feasts and festivals in Germany are at the time of full moon. 

 This, however, may be as much for the benefit of the light as the 

 lack of rain. In the U. S. Monthly Weather Review for October, 



' As Prof. Merriman's paper has not been seen, this must be regarded as 

 an independent discussion of the subject and not a reply to that.— H. A, H. 



RECENT BOTANICAL EXPLORATIONS IN IDAHO. 



BY- n. T. MACDOTJGAL, LAFAYETTE, IND. 



In various parts of the region occupied by the ranges, spurs and 

 foot-hills of the Rocky Mountains are large areas which have never 

 been explored by the naturalist. The species of the flora and 

 fauna of such regions can, to a great extent, be approximated by 

 a knowledge of the contiguous territory, especially if a similarity 

 of climate prevails, but in all cases every natural area of land, 

 such as a river, valley, or mountain range, gives to its forms of 

 plant and animal life certain differences from all forms found in 

 other localities. If the differences are of sufficient importance, 

 they will constitute new species, and in many cases whole groups 

 or genera peculiar to a certain region are found. 



The exploration of certain areas invariably brings 1o light 

 numerous undescribed forms of both plants and animals besides 

 affording valuable information on the distribution and variations 

 of known forms. 



At various times collections and observations on the flora of the 

 Rocky Mountains have been made by attaches of geographical 

 and geological surveys, and by the various parties engaged in the 

 exploration and survey of railroad routes across the continent, by 

 individual workers under the direction of the several divisions of 

 the U. S. Department of Agriculture, by representatives of various 

 scientific societies, and by collectors working entirely indepen- 

 dently. 



The amount accomplished in this way cannot easily he esti- 

 mated, but it may be suggestive to know that "The Systematic 

 and Alphabetic Index of New Species of North American Phanero- 

 gams and Pteridophytes," published in 1891 by Josephine A. 

 Clark, "Contributions from U. S. National Herbarium," Vol. I., 



