234 



N.-1 TURE 



[July 9, 1 S96 



tlevelopment in South-eastern Russia. Not only are there 

 rich deposits known to exist in immediate j-groximiiy to the 

 sea in the Transcaspian province, but an ininiense area of 

 petroleum-producing strata extends from the Crimea to the 

 Taman peninsula, and thence across the northern boundary of 

 the Caucasian ranf;e to Tetrovsk upon the Caspian, and many 

 centres of ptoducliun in these districts are now being opened up, 

 which must shortly come into keen competition with the Baku 

 industry. Already during many years oil has been extracted 

 from borings in the Kouban district, whence by means of pipe- 

 lines il is transported to a refinery at Novorossisk ; but these will 



Flo. 3.— The Temple of Zoroaster. 



be insignificant when (should all the reports be confirnied) the 

 wells at Groznaia and its neighbourhood are tapped, it being 

 considered that they will rival, if they do not surpass, Baku in 

 jirocUictiveness. It would appear that the beds are almost 

 identical in age to those of the Balachani-Sabountchi areas, and 

 it would be an interesting subject for future study to ascertain 

 if the line of petroleum productiveness to the north of the 

 Caucasus follows that of the depression which in a former 

 ].5eriod connected the waters of the Caspian w ilh those of the 

 Azoflf, the Black Sea, and the Mediterranean. 



The accompanying illustrations are reproduced from an 

 excellent series in Globus. W. F. Hume. 



A SEISMIC SURVEY OF THE WORLD. 

 'T'HE principal object of a seismic survey of the world is to 

 -*■ measure the velocity with which earthquake motion is 

 propagated through its crust, and possibly through its interior, 

 and from the resulting figures to give to astronomers and phy- 

 sicists additional data respecting its effective rigidity. 



It is the converse of the answer to a problem which in 1S89 

 was incidentally worked out by Lord Kelvin, who, assuming a 

 certain rigidity for our earth, determined the rate at which 

 vibrations were likely to be transmitted through the same, the 

 object of the calculation being to compare the result with that 

 obtained from observations on an earthquake which in that 

 year, originating in Japan, had been noted at many stations 

 n Europe. The feasibility of the proposetl undertaking and the 

 ]irobability of its yielding satisfactory results are based upon the 

 existence of observations of the following nature. 



For many years past astronomers, and those in charge of 

 self-recording magnetographs, have observed disturbances in 

 their instruments at varying intervals after the occurrence of an 

 earthquake in some remote locality. In 1S67, about seven 

 minutes after an earthquake in Malta, M. Wagner observed at 

 I'ulkova an oscillation of 3" in a level. One hour and fourteen 

 minutes after the great earthquake of Iquique on May 10, 1877 

 (effects due to which were observed by the writer in Japan), at 

 the same observatory M. Nyren noted oscillations in the bulb 

 of a level of 2" which had periods of 20 seconds. The late Dr. 

 E. von Rebeur-l'aschwitz repeatedly observed and obtained 

 records of earthquakes which had their origin at distances equal 

 to or more than one quarter of the earth's circumference frcmi 

 his observing stations. X'icentini, Agamennone, and others pro- 

 vided with instruments sensible to slight movements of the 

 earth's crust, have made similar records ; whilst the writer has 

 not only shared in contributing to this cla.ss of observations, but 

 on one occasion at least has obtained satisfactory photograms 

 of a disturbance originating at his antipodes. 



The conclusion which may be taken as well established by 

 these observations is that suitable apparatus jilaced in any part 



NO. 1393, VOL. 54] 



of the globe will record the movements due to severe earth- 

 quakes originating in any other portion of our globe, and there- 

 fore there is nothing unreasonable in saying that every observa- 

 tory throughout the world, if it were ecjuipped with proper 

 instruments, would be in a position to contribute to the know- 

 ledge of changes which are continually taking place, not only 

 beneath the land, but also beneath the ocean. 



Eor a person or a community to imagine that they reside in a 

 locality free from earthquakes, is one of the greatest of modern 

 fallacies. Although movements may not be felt, all places are 

 disturbed in a manner capable of being recorded very many 

 times per year. In addition to earthquakes the focus of which 

 may have been some thousands of miles distant, to the recording 

 of which (he present note is intended to draw special attention, 

 unfelt disturbances of a local origin may be recorded. Even at 

 places where shocks are unknown, excepting as rare events 

 recorded in ancient history, these may sometimes average two a 

 day. Other movements taking place beneath our feet are slow 

 diurnal tiltings, annual variations in the vertical, tremors of 

 probably two distinct characters, earth pulsations and elastic 

 vibrations. 



To designate all these movements, which vary in their periods 

 between the fraction of a second and twelve months, as '" earth 

 tremors," and an instrument to record them, a seismograph or a 

 tromometer, are evidently misnomers. Although a single instru- 

 ment may be obtained which will give information about each 

 of these movements, experience has shown that it is better 

 to have a particular instrument for a particular purpose. To 

 record rapidly recurring vibrations the inost sensitive arrange- 

 ment that is self-recording is, perhaps, a Perry tromometer, 

 which will detect the disturbance produced by a moving 

 train at the distance of a mile. To record slight changes of 

 level in a district, such, for instance, as may accompany changes 

 in barometric pressure, a bifilar or horizontal pendulum, which 

 is nearly as insensible to elastic tremors as a Perry tromometer 

 is to change of level, would be best. 



What is wanted for a seismic survey of the world is an instru- 

 ment that is sensible to the preliminary elastic tremors of an 

 earthquake, and then to the slowly recurring quasi-elastic gravi- 

 tational waves by which these are followed. For this purpose it 

 appears that our choice rests between some form of ordinary 

 pendulum apparatus, like that of Agamennone or \'icentini, 

 or some form of horizontal pendulum. Whatever form is selected, 

 each instrument must be similar and similarly adjusted. 

 If this is not the case, then at each station diflerent instru- 

 ments may commence to move with different |ihases of motion, 

 and the records for purposes of comparison are without value. 

 For example, an earthquake may originate at a known locality, 

 at a known time, and be recorded at twenty difi'erent observ- 

 atories in Europe, at each of which good time is kept, but at 

 each of which the recording instruments are ditTerent in 

 character. 



The result of the calculations based on these observations 

 have shown, in one instance at least, that the velocities of 

 propagation of motion from the origin to each of these stations 

 have varied between 2 and 20 km. per second. 



The cause of this apparent discrepaficy lies in the fact, that at 

 diflerent stations, during a disturbance having a duration of 

 perhaps several hours, the different instruments have commenced 

 to move with dilTerent phases of motion. This is a source of 

 error which has been thoroughly recognised by observers in Japan 

 for the last twenty years, and by timing the rate at which a 

 particular vibration has travelled between given stations, the 

 apparently conflicting results to which we are otherwise led have 

 been greatly reduced. 



When an earthquake is observed at stations far distant from 

 each other, it is no longer possible to identify a particular 

 vibration at these stations ; hut what can be done, is to note the 

 time at which the preliminary vibrations commence and the 

 interval which follows before the undulatory motiims appear. .So 

 far as observations have gone, the velocity of propagation of the 

 latter movements varies between 2 and 3 km. per second, which 

 is about the rate we should anticipate to be lound for motion 

 passing through the materials constituting the earth's crust. 

 The velocity for the former, however, appears to vary between 

 wide limits, 10 or 12 km. per second being about the average. 

 Becatise this rate of transmission is greater than that at which 

 motion could pass through glass or steel, the inference is that it 

 may possibly pass throiti^h our earth, and because it is variable 

 the idea suggested is, that the rate of transmission varies with 



