July i, 1887.] 



SCIENCE. 



\ 



portional to the square of the distance from the origin. Perhaps it 

 is needless to say that the observations had no such distribution. 

 But, after all is said, it must be admitted that there is much justice 

 in Professor Mendenhall's criticism of the isoseismals, and he cer- 

 tainly scores an important point. An earnest and conscientious 

 effort will be made to remedy the defect he has undoubtedly proven. 



As regards the ' areas of comparative silence,' I think they have 

 been too well established by the data in hand to be explained away 

 on the ground of defective testimony. They attracted attention at an 

 early stage of the investigation, and were at lirst thought to be due 

 to defective testimony ; but as the information increased, it was seen 

 that they were not so easily disposed of. Special inquiry was then 

 made, and the result was, to our thinking, a full confirmation of 

 their reality. 



In his criticism upon the method of computing the depth of the 

 focus, he proposes an argument which we anticipated would be 

 raised against it. He says, " As far as can be seen from the con- 

 tents of the paper, the result depends upon the unjustifiable as- 

 sumption that surface destruction is proportional to " the energy 

 per unit area of wave-front. I cannot admit that the paper implies 

 that assumption. But if he will permit me to substitute the word 

 ' effects ' for the word ' destruction,' then I will say that the result 

 does depend upon the assumption so modified, and stands or falls 

 with it. And, moreover, I hold that assumption to be not only 

 justifiable, but ne.xt door to an axiom. If our estimate of relative 

 intensities were to be derived solely from the destruction of build- 

 ings and chimneys by a force which in turn must be measured by 

 the maximum acceleration of the earth-particle in a horizontal plane, 

 our argument would indeed be in a pitiable plight. But we ought 

 not to be, and certainly are not, so limited. Other means of form- 

 ing an approximate estimate of relative intensity are abundant, even 

 where the destruction is little or nothing. Subject to local modifi- 

 cations, a great earthquake is bound to make itself felt somehow, 

 and in due proportion to its energy, whichever component, vertical 

 or horizontal, predominates. In the epicentral tract, brick build- 

 ings were few ; but there were plenty of wooden ones, and plenty of 

 intelligent men to tell what had happened. The best but by no 

 means the only inanimate testimony was furnished by the railroads 

 which cross this tract. They were like continuous lines of seismom- 

 eters ; and the men who repaired them had no difficulty in stating 

 where the road-beds were shaken up most, and where least, and 

 how the effects varied from mile to mile. 



What Professor Mendenhall really challenges, I infer to be, not 

 the theor)', but the competency of the data through which the theory 

 must be applied, if it can be applied. He appears to doubt the pos- 

 sibility of procuring such data ; but it seems to me that he overesti- 

 mates the exactions. He sees, indeed, that the vanishing of the con- 

 stant a dispenses with the necessity of making any absolute evalua- 

 tion of a single intensity, or even of the successive ratios between 

 intensities. All that we require is to find, if possible, where these 

 intensities var)' most rapidly along a line. It is analogous to trying 

 to locate, without the use of a level, the steepest point of a hill 

 whose profile is similar to our intensity curves. It cannot be done 

 exactly, but it can be done within moderate limits of error ; and I 

 have not much doubt, that, when Professor Mendenhall sees the 

 data, he will concede as much. It was distinctly stated in the paper 

 that the method was believed to be incapable alike of great pre- 

 cision and of great errors. 



But, though I cannot yield to his criticism on this point, I am still 

 greatly indebted to him for it. It is instructive in pointing out 

 sharply what treatment must be given to the data to enable readers 

 and investigators to judge of the validity of the method, and how 

 the facts must be marshalled. 



He also dissents from our inference that there were some facts in 

 Charleston which seemed hard to explain upon the assumption of 

 amplitudes of the earth-particle less than ten inches to a foot. This 

 was suggested as a maximum confined to a few spots, while the 

 mean amplitude was presumed to be considerably less. Let us ex- 

 amine this point. 



In all great earthquakes, those who have felt their violence near 

 the epicentrum have been impressed with, and testified to, an ap- 

 parently large amount of movement in the soil, — an amount to be 

 measured, so far as they could estimate, not by millimetres, but by 



inches, and sometimes even by feet. To verify these purely sensory 

 estimates was, of course, impossible ; but the circumstantial charac- 

 ter of the testimony seemed, in the absence of precise measurement, 

 to warrant the belief that the movements probably had about that 

 order of magnitude. When the seismograph was applied in Japan 

 to the measurement of the frequent but moderate shocks, and it 

 was found that an amplitude of a few millimetres would sometimes 

 crack walls and throw down chimneys, it was at once inferred that 

 the unmeasured estimates or guesses of the amplitude in the grander 

 shocks had been greatly exaggerated : for, the energy being pro- 

 portional to the square of the amplitude, it seemed needful to multi- 

 ply those already measured only a few times to obtain a destructive- 

 ness commensurate with that exhibited in the worst catastrophes. 

 There has been, therefore, a great change of opinion about these 

 large estimates among seismologists ; but I think it can be shown 

 that such estimates are not necessarily invalidated by the seismo- 

 graph. 



The intensity of a shock is not alone proportional to the square 

 of the amplitude, but also to the wave-velocity divided by the 

 wave-length. It is, I believe, a general fact that great amplitudes 

 of earthquake-waves are accompanied with great wave-lengths. 

 This does not follow from the accepted laws of wave-motion in 

 elastic solids, but is an independent fact, whose explanation must 

 go back to the nature of the originating impulses. Thus increasing 

 amplitude does not carry with it an increasing destructiveness in 

 so rapid a ratio as might at first be supposed. The displacement 

 is greater, but the time of displacement is longer. Again, the 

 amplitude diminishes as the wave moves on ; at least as fast as, 

 and probably faster than, the distance from the origin increases. 

 Let us, then, endeavor to make a comparison, rough though it must 

 necessarily be, between the larger amplitudes measured by the 

 seismograph, and those which may be inferred in localities shaken 

 by the Charleston earthquake with equal energy. I regard it as 

 improbable that the intensity of the most vigorous shocks measured 

 by the seismograph in Japan (so far as published) exceeded that at 

 Atlanta, Asheville, and Raleigh, all of which have been estimated 

 to exceed No. 7 in the Rossi Forel scale. If we take ten milli- 

 metres as the average amplitude of those places, we shall not ex- 

 ceed the higher ones recorded by the seismograph for shocks of 

 probably not greater intensity. The mean distance of these places 

 from the centrum is eleven and a half times as great as that of 

 Charleston. This would give an amplitude of about three inches 

 at the latter place, on the assumption that the wave-lengths were 

 equal to the Japanese, and that no energy was dissipated as the 

 waves moved on. The last assumption is certainly untrue, and, 

 whatever allowance may be made for it, must lead to a greater in- 

 ferred amplitude at Charleston. It does not seem to me that a mean 

 amplitude for the greater waves in that city, of three to four inches, 

 is too much, while local maxima may have been considerably 

 greater. The seismograph has not as yet tackled a first-class 

 earthquake in the vicinity of the central tract. 



Although I am still disposed to adhere, either wholly or in part, 

 to most of the propositions advanced in the paper referred to, I 

 must still acknowledge the high value of Professor Mendenhall's 

 criticism. It defines much more sharply the issues involved, and is 

 full of most useful suggestion. C. E. Dutton. 



Washington, June 23. 



Cyanhydric Gas as an Insecticide. 



Among the insect-enemies to plant-life, of which California has 

 received and is still receiving a full assortment from all parts of the 

 globe, the most formidable is the [cerya ptcrchasi, a coccid v^'hich, 

 instead of the hard shield that protects most of its congeners the 

 scale-lice, surrounds its egg-masses with a woolly fur that in many 

 respects serves even as a more efficacious protection. It has until 

 recently been supposed to have come from Australia ; but, accord- 

 ing to late researches of Professor Riley, it is to the Island of Mar- 

 tinique that we are indebted for this most pernicious insect. It 

 there infests the sugar-cane, and may readily have come in with 

 the canes often placed for drainage within the hogsheads of 

 raw sugar. Being apparently omnivorous, it has not been dis- 

 mayed by the absence or scarcity of its original plant - food. 

 Pine and cypress appear to be nearly as much to its taste as the 



