Jan. 21, 1875J 



NATURE 



223 



sun's heat become conductors ? Now, this can only take 

 place in the upper and rarer regions of the atmosphere, 

 since dense air is manifestly a non-conductor. We have 

 therefore in the upper regions of the air a conductor — 

 rare air— conveyed across the earth's lines of force by the 

 convection due to the sun's heat, probably with a very 

 considerable velocity. Notv, is it not possible that these 

 moving conductors may have currents generated in them 

 which *ill act upon the inagnet both directly and through 

 the earth ? As far as we ate aware no attempt has yet been 

 made to treat the question mathematically ; indeed, we 

 are hardly prepared for that at present, since we know 

 vety little about the convection currents in the upper 

 regions of the earth's atmosphere. 



We may perhaps, however, deduce the laws of the 

 Upper convection currents from what we know of the 

 lower currents. Now, there are several points of simi- 

 larity between the convection currents as we know them 

 and the daily magnetic variations. The first in order is 

 that noticed by Mr. Ba.^cendel), who observed a very 

 strong hkeness between the daily behaviour of the wind 

 and that of the magnetic decliaation. 



The iti'X-t is a resemblance between what we know takes 

 place near the equator as regards the magnetic decli- 

 nation and what we imagine must take place as regards 

 the upper convection currents. Sir E. Sabine has shown 

 that near the equator the diurnal magnetic change is of 

 an opposite character during the two halves of the year 

 reckoning from the equinoxes, so that it is only at or near 

 the equinoxes that the diurnal inequality might be 

 expected to vanish as it passes from the one phase to 

 the.other. Now, we should quite expect something of this 

 kind if the diUrnal changes were due to convection cur- 

 rents ; and just as the change which we might expect in 

 the convection currents of these regions on account of 

 the motion of the sun in declination would probably not 

 be gradual, but cf a hesitating or oscillatory character, so 

 Mr. J. Allan Broun has found from his magnetic obser- 

 vations at Trevandrum (page 180 of Dr. Lloyd's work) 

 that the magnetic change is not a gradual or regular one. 

 This is a very important remark, and if followed up by 

 a thorough discussion of the various tropical magnetic 

 observations, may be expected to throw much light on 

 the cause of the daily variation. 



The third point we would notice is a peculiarity in the 

 behaviour of the daily magnetic variation near the mag- 

 netic pole. 



" The observations of Sir Leopold M'Clintock in 

 1858 — 59, at Port Kennedy," says Dr. Lloyd, "have 

 enabled Sir Edward Sabine to throw further light upon 

 the laws of the diurnal variations. The declination at 

 Port Kennedy is N. 136° W. ; while that of Point Barrow 

 is N. 41° E. The north poles of the needles at the two 

 stations, which are at opposite sides of the earth's mag- 

 netic pole, thus point in opposite directions. Now, 

 when disturbances are removed, the observations gave 

 the greatest deflections at 8 \m. and 2 P.M., as in other 

 places. But they showed, further, that the positions were 

 referred in both to the mai^nctic mtiidian of the place, 

 and not to \h^ astronomical ; the deviations of the magnet 

 at 2 P.M., for example, being in both places to the left of an 

 observer looking towards the magnetic pole at each place, 

 and therefore geographically in opposite directions." 



Now, meteorologically, the north magnetic pole is not 

 far from the pole of greatest cold, and we might, perhaps, 



expect on opposite sides of the pole [to find the upper 

 convection currents going inapposite directions. If this 

 be the case,'"and if the daily variation be due to those 

 currents, then we might also expect a magnetic be- 

 haviour such as was deduced by Sir E. Sabine from the 

 observations of Sir L. M'Clintock. 



We think, in fine, that the behaviour of the daily 

 variation lit the tropics, at middle latitudes, and near the 

 magnetic pole, is not inconsistent with the hypothesis 

 that sUch variation is due to convection currents. But if 

 this hypothesis be true, it cannot be limited to the daily 

 variation. We know very well that the currents of the 

 earth's atmosphere often present great irregularities, and 

 that these irregularities are especially prevalent at the 

 equino.xes. Now, we have a precisely similar peculiarity 

 m magnetic changes. These are frequently irregular, and 

 their irregularities are greatest at the equinoxes. In 

 proof of this we extract the following table from Dr. 

 Lloyd's work : — 

 Annual variation of the mean disturbance at Dublin. 



The next point to which we would allude is a simi- 

 latity between the secular variation of the meteorology 

 and magnetism of the earth. Mr. Baxendell, we think, 

 was the first to point out that there is a change in the 

 convection currents of the earth, depending on the state 

 of the sun's surface with regard to spots ; and Mr. Charles 

 Meldrum has followed with the very interesting and im- 

 portant announcement that we have most frequent cyclones 

 in the Indian Ocean during years of maximum sun-spots ; 

 and finally, M. Poey has shown that there is a similar 

 correspondence between sun-spots and the hurricanes ot 

 the West Indies. In fine, we have here an intimate con- 

 nection between solar and terrestrial meteorology. But 

 we have also a connection between sun-spots and mag- 

 netic disturbances ; and Sir E. Sabine was the first to 

 point out that during the years of greatest sun-spots we 

 have the greatest disturbance of terrestrial magnetism. 

 Now, may not the increase of magnetic disturbance be 

 due to the increase of meteorological disturbance which 

 the sun somehow produces, the upper convection currents 

 influencing the magnet in the manner above stated ? 



It is probable, however, that some will raise the fol- 

 lowing objection to this hypothesis. When there is a 

 great magnetic storm or disturbance, this takes place 

 simultaneously and abruptly throughout the whole earth ; 

 now, how can this be the result of a meteorological com- 

 motion 1 



We would reply to this objection that magneticians 

 have begun to recognise two sets of disturbances. 



When the writer of this notice was at the Kew Observa- 

 tory, this was forcibly brought before him. There are dis- 

 turbances of a rounded character, and there arc others 

 which are exceedingly abrupt ; and we think that Senhor 

 Capello has shown that these rounded disturbances are 



