86 



NATURE 



\_Afay 26, 1887 



tinued till we reached the line, and the single observation 

 that I got of high cirrus in 1° N. lat. showed an easterly 

 current at that level. The above results are very curious, 

 for the usual idea is that the south-east trade on crossing 

 the equator becomes a south-west wind under the influ- 

 ence of the earth's rotation ; but here, though the surface 

 wind appears to conform to the usual rotation of the air 

 round a low pressure, the upper currents maintain the 

 vertical succession of the southern hemisphere. 



After crossing the equator ourselves, the wind turned to 

 south-by-east, or south-south-east; and as far as 18° S., 

 beyond which we need not follow them, the lower currents 

 were either identical in direction with the surface winds, or 

 else a very little more easterly — that is, they followed the 

 normal vertical succession of the southern hemisphere. 



I have endeavoured to show all this in a diagrammatic 

 form in Fig. i, where the full arrows denote the surface 

 wind ; the broken-lined arrows the direction of the low 

 or middle-level clouds ; the dotted arrows that of the 

 highest cirrus. A couple of arrows to denote the typical 

 succession over an extra-tropical south-west wind are 

 inserted on the upper left-hand corner of the figure. 



The observations in the Indian Ocean have also been 

 published in Nature, vol xxxii. p. 624, and vol. xxxiii. 

 p. 460. They were taken during two trips — one from Aden 

 to Cape Lewin, in February, the other from Cape Lewin 

 to Colombo, in the same month of the following year. 



SO'.N. 



30'. S. 

 SO'U/ 40' 30° 20" 10" 10° ZO'EI. 



Fig. I. — Surface and upper currents in the Atlantic. 



This is the season of the north-west monsoon. The 

 results were very accordant, and showed conclusively that 

 in vertical succession the upper currents over the north- 

 east monsoon were always more from the east than the 

 surface wind, as is usual in the northern hemisphere. 

 No high cirrus was ever observed in this part of the 

 ocean. 



In the north-west monsoon, between the equator and 

 the doldrums, low and middle currents came from north 

 ■or north-east, and the only two observations on the 

 highest cirrus showed an easterly current at that level. 



In the south-east trade, the low or middle currents 

 were generally about from the same direction as the sur- 

 face wind, or else a little more from the east. This is the 

 normal succession of the hemisphere. Thus we see an 

 ■extraordinary analogy to what occurs in the Atlantic. 

 When the north-east monsoon is drawn across the 

 •equator, towards the low pressure of the doldrums, the 

 surface wind seems to pick up the westerly component of 

 the earth's rotation in the usual manner ; but the upper 

 currents retain the vertical succession of the northern 

 hemisphere. 



The results of all the observations are given in a dia- 

 grammatic form in Fig. 2. It will be remarked how few 

 observations there are of cirrus ; this is because it is so 

 rarely seen in those latitudes. It is certain that, on the 



southern and western edges of the great anticyclone 

 which controls the south-east trade, the highest currents 

 would have come more or less from the north-west, as is 

 marked at Mauritius. 



I have unfortunately got no equatorial observations 

 from the Pacific, but they would most likely be very 

 similar to those in the Atlantic. 



However, the upshot of all those which we have just 

 described is to show that when the trades or monsoons 

 meet they do not interlace, as has been suggested by 

 Maury, but that the upper winds combine in a generally 

 easterly current, and probably diverge only slightly pole- 

 wards on either side. I am unable to form any opinion 

 as to the velocity of this current. 



I have often been asked how far all this bears on the 

 remarkable distribution of dust from Krakatao, and the 

 answer is very simple. The great explosion of Krakatao 

 took place on August 26 ; that is to say, during the south- 

 west moonsoon of the Indian Ocean. The distribution 

 of th2 surface and upper currents is then very different 

 from that in our last diagram. The south-east trade, 

 with its upper easterly currents, extends all over the 

 Indian Ocean, and Malaysia south of the line, including 

 Krakatao, marked K in the figure. North of the equator 

 the surface wind turns to the south-west, but we know 

 nothing of the motion of the upper currents. 



Now, assuming that the blue sun and unusual twilight 



30 o 

 dO'E 40° 70° 80° 90° 100° liO' ltO°E 



Fig 2.— Surface and uppe: currents i i the Indian Ocean in Febru.ary. 



glows were equally the product of volcanic dust as the 

 fall of actual ashes, the sequence of the first appear- 

 ance of these phenomena in different parts of the world, 

 dating from August 26, was as follows : — 



On August 26, the day of the eruption, ashes, lofty 

 haze, or red twilights are reported south of the equator, 

 nearly 20^ of longitude west of Java, from one station in 

 Sumatra, just north of the line, and from near Formosa. 



Next day, the 27th, similar phenomena were reported 

 from many stations in the Indian Ocean, south of the 

 equator, as far west as Mauritius and the Seychelles ; 

 while north of the line, strange appearances were reported 

 for the first time in Borneo and Ceylon. 



On the 28th, the haze and abnormal glows had extended 

 to Natal on one side and Japan on the other. 



No important extension of the area is reported on 

 August 29 ; but by the 30th unusual coloration of the sky 

 is reported from various parts of the South Atlantic, 

 and Guiana, and, what is specially noticeable, from about 

 the Cape Verd Islands, north of the equator. 



On the 31st, another station in Brazil, and also a West 

 Indian island, report a strange look about the sun or sky ; 

 while on September i the same was noted at Guayaquil, 

 on the west coast of South America ; and in a quite un- 

 expected locality, far away from there, — New Ireland. 



September 2 was characterized by an outburst of 



