32 



SCIENCE. 



[Vol. XVI. No. 389 



Planetary Hypotheses. 



If the moon, only 250,000 miles away, has no marked in- 

 fluence on our weather, what must we think of the effect of 

 the planets, millions of miles away? It is no wonder that 

 at least one of these prophets, after giving the whole subject 

 careful study, was forced to abandon the planetary hypothe- 

 sis for the lunar theory. There is nothing which shows the 

 utter absurdity of these planetary theories more forcibly 

 than the introduction of the hypothetical planet Vulcan. 

 This is needed in order to have a body revolving around the 

 sun frequently enough to make his position relative to the 

 earth coincident with our numerous storms. One of these 

 prophets, an American, thought he saw Vulcan passing 

 across the sun, and published a careful computation indi- 

 cating, that, according to Le Verrier's orbit, the planet 

 should have been exactly at that point; but, unfortunately 

 for this hypothesis, it was shown by Professor Proctor, that, 

 owing to a slight inaccuracy, this computation was wrong, 

 and that this prophet, if he saw Vulcan at all, must have 

 seen it directly through the sun, on the opposite side from 

 the earth. Granting that there is a planet only 8,000,000 

 miles from the sun, and about 85,000,000 from us, is it not 

 perfectly plain that its influence on terrestrial weather would 

 be most completely overshadowed by the all-powerful sun 

 behind it? 



It is not a little remarkable that these prophets are entirely 

 disagreed as to how this planetary effect is produced. One 

 would have our storms and tornadoes coincident with the 

 equinoxes of the planets, another with their oppositions and 

 conjunctions; and so on. It is easy to see, that, under these 

 circumstances, no two of these prophets agree on the dates 

 of storms, but they are distributed quite uniformly for about 

 half the days in the year. How does it happen, that, though 

 these dates disagree, all these prophets are perfectly satisfled 

 as to the exact fulfilment of their predictions? This is very 

 simple to explain; for the man who predicts a storm on the 

 1st of the month, for example, verifies by a storm, say, in 

 Illinois, while the one who has put his storm on the 3d of 

 the month verifies by the same storm, which has now moved 

 to Maine. There is just one crucial test by which we may 

 satisfy ourselves of the worthlessness of these theories. It 

 lias been outlined in the last chapter. Let these prophets 

 make a careful study of all the influences they can muster, 

 and put down, no matter whether for the past or the future, 

 the dates when they would expect the worst storms, and also 

 the dates of fewest disturbances, then take the whole extent 

 of this country, and establish the dates of most and least at- 

 mospheric disturbance. A comparison of these dates would 

 quickly prove the value of such predictions. It is needless to 

 add, that frequent and continued attempts to obtain these 

 dates from at least four of these prophets, and to get any 

 one of them to agree to this comparison, have lamentably 

 failed. 



There are not a few people who put great faith in such 

 predictions, though a moment's thought would show how 

 preposterous the claim is. For example : the Louisville tor- 

 nado, on March 37 of the present year, was heralded as a 

 perfect verification of a prediction for storms from March 28 

 to 30, and pains were taken to spread this fact from Maine 

 to California. Suppose some one in the tornado district had 

 read this prediction on March 27, and put faith in it: would 



he not have been misled? Again : if some one in Maine had 

 read the prediction, would a storm in Kentucky apply to his 

 locality? It is so easy to make a storm prediction, and so 

 easy to verify it if one is allowed his own way, that there is 

 no immediate prospect of silencing these prophets; but it is 

 to be hoped that our citizens will study this matter for them- 

 selves, and before long obtain right views. It is plain that 

 such a prediction made years beforehand can have no influ- 

 ence on right-thinking persons, for we know that it is im- 

 possible to predict the weather with certainty for even 

 twenty-four hours. 



Since 1872 it has been known that tornadoes and severe 

 storms occur in the south-east quadrant of a depression sys- 

 tem as it traverses the country, and in the history of the Sig- 

 nal Service frequent predictions of such storms have been 

 made. A great deal of discussion has arisen as to the possi- 

 bility of extending this system, and of giving ample warn- 

 ing of these outbursts. The most that can be said at present 

 is, that the occurrence of such a storm is exceedingly rare ; 

 and in a very small space, while we may be able to indicate 

 a region of several thousand square miles where such local 

 outbursts may be expected, yet little more than this can be 

 hoped for. People living in such districts, when they hear 

 of the prediction, should not be disturbed, but simply take it 

 as a probable occurrence at possibly one or two places, and 

 in any particular locality should be guided by tfle appear- 

 ance of clouds and other threatening signs which have be- 

 come familiar. In fact, the question frequently arises as to 

 whether it would not be better to omit such a prediction en- 

 tirely ; but if the right view be taken of it, that it is a warn- 

 ing to look out, and not a positive statement, no one should 

 be unduly disturbed. 



There are times when there seems to be an unusual 

 amount of electricity present in the atmosphere, and when 

 these severe storms occur without presenting any indication 

 whatever on our maps or in our reports. It is impossible, 

 from our present telegrams and knowledge of these storms, 

 to make any predictions in such cases, though we may hope 

 that in the future we may have a clearer idea of disturbed 

 conditions at one or two thousand feet above the earth, 

 which will enable better predictions. Such storms are not 

 very severe as a general thing. A storm region like that in 

 Kentucky on March 27, 1890, is plainly indicated on our 

 maps, and predictions of severe local storms were sent all 

 through that region nearly twelve hours in advance. 



There has been a gradual development in these predictions 

 as the conditions have become more familiar. One of the 

 later attempts was made in 1884, and in this case the whole 

 country east of the 102d meridian was divided into eighteen 

 districts, and private predictions were made each day during 

 the tornado season as to whether or not a tornado would 

 occur in any district. The claim was made that in this case 

 97 per cent of the predictions were successful, but a serious 

 fallacy in these attempts was soon pointed out. To say that 

 on any day in New England, for example, there would be 

 no tornado, was no prediction at all ; for only under most 

 extraordinary conditions, occurring once in three or four 

 years, are any tornadoes experienced there. Several verifi- 

 cations of these predictions according to mathematical prin- 

 ciples gave from 13 to 20 per cent of success. This does not 

 indicate, however, the measure of skill that has been at- 



