404 PROFESSOR C. MICHIE SMITH ON OBSERVATIONS ON 



this would be the case. The experiments made by Dr Oliver Lodge {Nature, 

 vol. xxix. p. 612) seem quite insufficient to found any theory upon, inasmuch 

 as they deal simply with the effect produced on a mass of very fine particles of 

 water in an enclosed space by a highly electrified point. 



The presence of abundance of vapour seems to be explained, naturally 

 enough, by the setting in of the moist monsoon currents in the upper parts of 

 the atmosphere; or at least by the conflict between the north-east and south- 

 west monsoons, which was apparently begun by that time* It is by no means 

 impossible that the Krakatao eruption may have had some influence on the 

 direction of these currents, for we have proof of the immense displacement of 

 the air caused by the eruption in the barometric waves which were traced three 

 times round the earth. Further, the ascent of a heated column of air and 

 vapour over the volcano would tend to produce an area of low pressure round 

 it, and thus to set up a cyclonic influx of air from other places, and it is quite 

 possible that this influx is indicated in the oscillations of the barometer shown 

 in the accompanying diagram, the sudden cessation of the cause of inflow after 

 the eruptions ceased, causing a series of decreasing waves — direct and reflected. 

 The eruption may also have had something to do with the electrical state of 

 the air, for we know from observations made on the spot, that much electricity 

 was generated by the eruption, and Professor Palmieri's observations show 

 that the material ejected from Vesuvius is negatively electrified. A hypothesis 

 connecting the phenomena with the eruption at Krakatao in this way would not 

 be liable to the same difficulties regarding the velocity of propagation as one 

 which involves the actual transport of dust, for the delay at one place relatively 

 to another might be caused by the absence of a sufficient quantity of vapour. 

 The theory which I have suggested would also account naturally for the second 

 appearance of the green sun, the interval being due to one of the familiar lulls 

 in the contest between the two monsoons. On the dust theory, it seems very 

 difficult to account for this second appearance. 



At one time I was inclined to think that the theory which made the 

 phenomena depend on cosmic dust was an extremely probable one, and that 

 the dust formed nuclei, about which the vapour condensed in the manner 

 shown by Mr Aitken. Various observations, however, seem to negative this. 

 It seems hardly possible that vapour was condensed in any quantity, else the 

 absorption of solar heat would have been considerable. Observations made 

 with the black bulb thermometer in vacuum are not very satisfactory, yet when 

 carefully made with the same instrument under similar circumstances, they 

 have some value. The Madras observations show that the readings were 

 above the average for the month, and that on the 23rd, one of the days of 



* Sec Mr Levinge's remarks, ante, p. 396, and the extract from the log of the " Nizam," p. 399. 



