22 



SCIENCE. 



[Vol. X. No. 231 



of fact, the sun does not heat up a Hmited portion of the earth. 

 Its rays shine with equal intensity over i.ooo miles from east to 

 west. It has also been shown that this heating of the surface 

 does not ascend more than a few inches in the air. One 

 of the strongholds of the theorists is unstable equilibrium ; but 

 right here we find two seemingly contradictor)' statements. On 

 p. 51 of Professor Ferrel's book, ' Recent Advances in Meteorology,' 

 there is a suggestion that this state (unstable equilibrium) is 

 brought about whenever there is a less diminution of temperature 

 with height in an ascending column than in neighboring portions 

 of air. On p. 328 of the same volume, however, the idea is given 

 that this same state may be produced if there is an abnormally 

 great diminution of temperature with height. It would seem as if in 

 both these instances, even if there were a tendency to this state, air 

 would flow in at all times from surrounding regions, and instantly 

 relieve the condition. This relief would be afforded the more 

 rapidly, the less the friction. However, the error here is farther 

 back. We cannot suppose that the atmosphere is either quiescent 

 or flowing in a current having a uniform velocity in all its layers, 

 to the height, say, of 15,000 feet. The fact is admitted that there is 

 a uniform acceleration in the different strata as we arise ; so that, 

 ■even if an upward movement should begin, a few hundred feet would 

 ■destroy all vertical tendency. As a matter of fact, when we con- 

 sider the actual conditions under which solar radiation acts at a 

 storm-centre, we see that this unstable state could not be formed. 

 At a storm-centre clouds cover the earth's surface, and prevent 

 all abnormal conditions from great heat. Balloon-ascents have 

 ■shown uniform temperatures up to the top of the clouds. 



The theoretical computations of the velocity of the upper air strata 

 ■do not correspond with the actual movements recorded. On p. 259 

 Professor Ferrel gives the velocity of the current at the height of 

 16,000 feet as 26 miles per hour in the middle latitudes of the 

 United States. 



^On Mount Washington, 6,300 feet in height, the velocity when a 

 low area passes is 53 miles per hour, and when a high area passes 

 it is 21. The velocity of the low areas near Mount Washington is 

 34 miles per hour. This would indicate that the ' power ' of the 

 storm must be below 6,300 feet, since it is admitted that its pro- 

 gressive motion is due to the movement of the strata where it ex- 

 ists. It may be safely said that a height of less than 6,000 feet for 

 the centre of disturbance would be fatal to a great many of the pres- 

 ent theories of storm-generation. 



Formerly it was said, that, owing to friction with the earth's sur- 

 face, the upper part of the storm must be in advance of the lower ; 

 but it is certain that such a state of things could continue only a few 

 minutes, for the upper portion of the storm would be rapidly sepa- 

 rated from the lower. Professor Ferrel, on p. 260 of the present 

 volume, explains this difficulty by suggesting that the upper part of 

 the storm is continually re-forming itself, and that there is no actual 

 transferrence of air. I hardly think that this suggestion will be 

 accepted. It seems to me our storms would behave differently if 

 it were true, and certainly our synoptic charts do not give any clew 

 to such re-formations of the upper part of the storm. It seems to 

 me this later theory destroys the continuity of the ascending cur- 

 rent and the essential features of unstable equilibrium. One of the 

 most difficult phenomena to explain is the fall of rain at a distance 

 of 300 and more miles from the storm-centre. If we suppose the 

 ascending currents are at the centre of the storm, then rain should 

 fall at that point. Professor Ferrel, at p. 266, advances the novel 

 idea that the rain is formed in or carried to the upper currents, and, 

 as these are more rapid than the storm, it must fall in advance of 

 the storm. I do not think this theory takes sufficient account of 

 the facts. Let us suppose the raindrop carried to a height of 7,200 

 feet : obsen'ations irf balloons show that rain very rarely occurs 

 above that height, and that the ' power ' of the storm is at 5,000 

 :feet. We may consider the velocity of the current at 7,200 feet 15 

 miles per hour greater than at 5,000 feet : the drop would fall at 

 about 10 feet per second, or would reach the earth in 12 minutes ; 

 and hence, if it had been carried in the upper current during this 

 time, it would have fallen 3 miles in front of the centre, instead of 

 300 or more. As a matter of fact, since the currents below 5,000 

 feet are very much slower than above that height, any acceleration 

 would be entirely overcome, and from these principles the drop 



would actually fall back of the centre. On the continent of Europe 

 the bulk of the rain falls at the rear of the storm. 



To my mind, howe\-er, theoretical meteorology most signally fails 

 in its attempts to explain our more violent storms and tornadoes. 

 That the sun's heat could start a vertical current which, with the 

 condensation of moisture in the upper air, would give rise to winds 

 of 200 or 300 miles per hour, seems incredible. The attempt to 

 meet the difficulties by suggesting ' great contrasts of temperature,' 

 ' meeting of warm southerly with cold northerly winds,' ' cool air 

 overrunning warm,' ' warm air o\'errunning cool,' etc., does not 

 seem at all satisfactor)'. As long as it was supposed that tornadoes 

 occurred at the centre of a low area where it was thought there was 

 an ascending current, the theory seemed plausible ; but when it 

 was clearly shown, in March, i S84, that tornadoes do not occur at 

 a low centre, but 400 or 500 miles to the south-east, it became 

 necessar)' to explain this. It seems to me that all attempts to eluci- 

 date this subject have merely served to lighten the darkness with- 

 out removing it. 



There is no space left for minutely examining the great super- 

 structure built on what seem weak foundations. It seems as though 

 the first and most important step is to remove the slur cast upon 

 this science by those who are qualified to know its weakness. Let 

 our theorists bend every energy to establish some fundamental 

 proposition-, either by actual experiment in the laboratorj' or by in- 

 vestigation in nature's laboratory at the spot where the ' power ' of 

 the storm manifests itself. It seems to me the recent attempts of 

 Weyher in France to demonstrate the existence of this ' power,' by 

 means of a rapidly revolving fan at some distance abo\-e water or 

 grain, show the great need of further proof. These experiments 

 show what might be if only there were an enormous fan in the 

 upper air, but where is the fan ? Must we not conclude that the 

 true explanation is now farther off than before, and certainly much 

 farther from the present theories. H. Allen Hazen. 



Washington, July i. 



Determination of the Depth of Earthquakes. 



The report of Captain Dutton and E\'erett Hayden on the 

 Charleston earthquake {Science, ix. p. 489) is undoubtedly a very 

 valuable addition to earthquake literature. There are two or three 

 points, however, to which I wish to draw scientific attention, in the 

 hope that investigation hereafter may clear them up. 



Perhaps the most interesting and important point in the report is 

 their method of determining the depth of earthquakes. The authors 

 first review rapidly other methods. Mallet's method — by protract- 

 ing the lines of emergence back to their meeting-point — they dis- 

 miss as too uncertain. Seebach's method — used in the earthquake 

 of Central Germany in 1872, which depends on the law of decreas- 

 ing velocity of the emergent wave — they also dismiss, because 

 the times of arrival at different points cannot be determined with 

 sufficient accuracy, on account of the different velocities of the two 

 different kinds of waves, normal and transverse. In place of these 

 methods they propose what they claim to be a wholly new one, 

 founded on the law of decrease of intensity; i.e., of decrease of 

 the shock-motion or motion of the earth-particle, or, in other 

 words, the wave-height or amplitude.' They show by mathemati- 

 cal discussion that the place of the maximum rate of decrease of 

 intensity bears a fixed relation to the depth of the focus ; viz., as 

 I to Vz- Upon this basis they estimate the depth of the focus to 

 be about twelve miles. In Fig. i, which we reproduce from their 

 report, the fall of the double-curved line represents the decreasing 

 intensity. The place of most rapid fall, i.e., where the cun-e changes 

 from convexity to concavity, is the place of most rapid decrease of 

 intensity. This place was quite distinctly marked. It was about 

 seven miles from the epicentrum. 



We wish now to draw attention to the fact that this method does 

 not differ verj- greatly from, and perhaps is not an improvement 

 upon, another method suggested by INIallet in his ' Report to the 

 British Association, 1858,' p. 102, though not used in his discussion 

 of the Neapolitan earthquake of 1857 ; viz., by means of what may 

 be called ' the circle of principal disturbance.' This method is 

 mentioned and explained in my ' Elements of Geology,' p. 117. The 

 authors seem to have overlooked it. 



with constant ^ 



^-length, intensity « amplitude. 



