other in that of the British Islands, the former being stationary, 
_ the latter in motion ; and as it is said now to be Siberia, its 
- motion must have been from west to east. On the other hand, 
we find this motion described as being in the opposite direction, 
--viz., from east to west. Sir George Airy, in his treatise on 
Magnetism, p. 93, remarks “that at Greenwich the dip and 
total force are diminishing. Interpreting these by the remarks, 
it would seem that the magnetic equator is approaching above 
Greenwich, or the north magnetic pole is receding from Green- 
wich ; and remarking also the westerly change in direction of 
north magnetic meridian, from the sixteenth century to the year 
1824, and its subsequent easterly motion, it would seem that the 
north magnetic pole has rotated round the terrestrial pole in a 
small circle from east to west, and, having passed the point where 
its westerly azimuth, as viewed from Greenwich, is maximum, is 
still continuing its course in that circle.- It seems probable that 
in the fifteenth or sixteenth century it was situated between North 
aig and Spitzbergen. It is now north-west of Hudson’s 
ay. 7 
The north magnetic pole, which Sir E. Sabine supposes to 
have been in Halley’s time in the meridian of the British Islands, 
would appear to be the same which Sir G. Airy says was 
probably in the 13th and 16th century in a meridian between 
North Cape and Spitzbergen ; yet the pole referred to is in one 
case said to be now in Siberia; while in the other it is said to be 
to the north-west of Hudson’s Bay; but it cannot at the same 
time have gone to the eastward and also to the westward. 
In the two accounts there is a discrepancy, but perhaps this is 
apparent only, and some of your readers may be able to show 
how the accounts can be reconciled. x 
Height of Thunderclouds 
A FEW days ago I had an opportunity of estimating the height 
of a bank of thunderclouds, an account of which may interest 
the readers of NATURE. 
I was camped at Gurpur, a place some eight miles from and 
within sight of the sea, with an elevation of about 480 feet. The 
evening was fine, and the horizon to westward remarkably free 
from haze, so much so that when the sun dipped it was still too 
bright for the naked eye to bear. Some fifteen or eighteen 
miles to eastward a heavy thunderstorm was raging, and the 
Western Ghats were shrouded by immense masses of cumulus, 
which, piled up to an enormous height, and rosy with the beams 
of the setting sun, formed quite a study for an artist. 
Having been in the jungles for three or four weeks, I noted, 
in order to get correct time, the moment the sun disappeared 
beneath the horizon. This was, by my watch, six minutes past 
six o’clock. A few minutes subsequently I noticed the earth- 
shadow creeping up the clouds to eastwards, its edge being singu- 
larly well-defined by the contrast of the cold grey beneath and 
the warm colouring above. Struck by the slow progress of the 
shadow, I timed it, and found that at seventeen minutes past 
six the last tinge of pink faded from the highest point of the 
cumulus, and at nineteen and a quarter minutes the cirrhi float- 
ing above the storm lost their hue, thus giving eleven minutes for 
the former and thirteen and a quarter for the latter. These times 
reduced and corrected for latitude (130° N.), give the respective 
approximate heights of the clouds as 14,075 and 25,590 feet, or, 
adding height of observer, about 2°75 and 4°93 miles. 
To be on the safe side, but 10’ of arc have been allowed for 
the eastward position of the clouds. The nature of the observa- 
tions of course renders correction for refraction unnecessary, so 
that the above figures are well within the mark. 
I believe that in the tropics cumuli attain a considerably 
greater elevation than is generally believed. In 18641 was on 
board a vessel in lat. 2° 53’ N., long. 10° ‘47’ W., when there 
were constant flashes of forked lightning visible among detached 
clouds directly overhead, yet not the faintest growl of thunder 
was heard by anyone on board, although a dead calm prevailed 
at the time. This fact I can only attribute to the combined 
effect of the immense altitude of the clouds and the consequent 
rarity of the air. E. H. PRINGLE 
Mangalore, South Canara, Nov. 2 — 
PERIODICITY OF RAINFALL 
HiAvine been working at the above subject for the 
last ten years, it occurs to me that a brief record 
of my failures and successes will form an appropriate sup- 
NATURE 
_ 143 
plement to the important article by Mr. Lockyer in 
NATURE for Dec. 12. 
Meteorologists have been hunting for a Saros through- 
out the present century. Among them, perhaps, the 
most devoted to the subject were Lieut. George Mac- 
kenzie, author of “ The System of the Weather,” and 
Luke Howard, whose “Cycle of Eighteen Years in the 
Seasons of Britain” is a well-known work. 
What little I have done in the subject is briefly told. 
-Almost immediately after commencing the collection of 
British rainfall statistics, which has now reached a com- 
pleteness excelling that of any other country, my atten- 
tion was naturally drawn to the question of periodicity.’ 
Kaowing, however, something of the care requisite to 
obtain long series of observations strictly comparable, I 
waited five years before printing anything bearing upon 
it; in 1865, however, I prepared and published * the fol- 
lowing table for fifty years, based upon the mean of 
continuous records in different parts of Great Britain :-— 
Tasre I.—Mean Depru oF RAIN AT TEN STATIONS, 1815—1864 
Year. | Depth.|| Year. | Depth.'| Year. | Depth. Year. | Depth. | Year. | Depth. 
1815 | 27°12 || 1825 | 26°57 |] 1835 | 28°56 1845 | 27°87 || 1855 | 23°37 
1816 | 29°26 || 1826 | 23°76 |} 1836 | 33°49 || 1846 | 29°57 || 1856 | 25°89 
1817 | 29°73 |] 1827 | 29°53 || 1837 | 24°54 || 1847 | 25°80 1857 | 25°70 
| 
| 
1818 | 30°34 || 1828 | 33°02 || 1838 | 27°1x 1848 | 35°98 || 1858 | 22°79 
1819 | 30°46 || 1829 | 28°70 || 1839 | 32°27 || 1849 | 28'5z || 1859 | 28°53 
1820 | 24°53 || 1830 | 30°83 || 1840 | 24°67 |] 1850 | 26°35 || 1860 | 33°34 
x82 | 29’92 || 1831 | 32°28 || 184x | 33°52 1851 | 26°70 || 186x | 26798 
1822 | 26°63 |] 1832 | 26°20 || 1842 | 25°53 || 1852 | 35°53 || 1862 | 30°37 
182 31°09 || 1833 | 20°72 || 1843 | 30°40 |] 1853 | 27°38 || 1863 | 26°93 
1824 | 30°9t |] 1834 | 24°52 || 1844 | 23°72 || 1854 | 22°38 || 1864 | 22°rr 
Mean,| 28°999 | 28'512 28'280 | 28°607 26°60r 
I also called attention to two features in this table, 
which strongly tend towards the confirmation of Mr, 
Meldrum’s views, viz. :—(1) that the wettest years are 1836, 
1841, 1848, 1852, and 1860; (2) that of these, all but two | 
form a 12-year period, viz., 1836, 1848, 1860, to which we 
may now add 1872+; (3) that the dry years were 1826, 
1834, 1844, 1854, 1855, 1858, and 1864; (4) that of these, 
all but three form a Ic-year period, viz,, 1834, 1844, 1854, 
and 1864. 
All this looked very satisfactory ; but, to make assurance 
doubly sure, I determined to make up a longer period. 
This I accordingly did; and the approximate fluctuation 
of annual rainfall during one hundred and forty years, 
viz., 1726 to 1865, will be found in the British Association 
Report for 1866, page 286, e¢ seg. These values were 
converted into ratios, and, subsequently, those for the 
years 1866 to 1869 were added, and the table was given 
in the following condensed form in an article on the 
“Secular Variation of Rainfall in England since 1725,” 
Taste II,—Ratio or THE Fart or Rain tn Eacu YEAR SINCE 1725 TO THE 
MEAN FALL oF Sixty YEARS, ENDING WITH 1869 
———— Oe 
Vear.|1720 1730/1740 1750 1760, 1770 1780 1790 1800| 1819) 1820 | 1830 | 2840 lr850) 1860 
° 88] 65 | 6) 70) 108 | 75, 86 g0| 100} 92|11r| 8&9 gt! 122 
ee} 71| 58 108} 87; 79 79 105 96 | 98 109 | 108 | 128 | 88 go 
ay 83) S| 82) 7x 11x | 13%, 117| gr | 97) 100 | 98 | ox | 138) 107 
3 71| 60 | 87) 1x8) 113 | 93 86) 77 | 92\ 117 | 106 | r10 | ror) 8g 
4 114} 89 76, 101) 329 | 96 x04) 85 | 92) 117 | go] 85| 74) 73 
5 to2| 80 | 83 82) 12 77 84 75) 99) 96) 9g 7 | 88 10g 
6 109] tor) 70 | 100, 77) 107) 127 83) 96 | 107) 77 | x18 | 108 93) 115 
7 102} TIO} 95 | 93| 91; 89 | 96 106 94 | 100) 102) 87] go} 97 103 
8 | 109] 70) 84! 128) 102 65 88 90/| 102) 120) go| 130} 80) roo 
9 | 97 80) 59 |} Sr 86} 83 | 116 106 88 99) 102 | 107 | 98 | to2 104 
a |e ed a ag = Ss 
Mean) ... | 89’9 70°) 85°5 or 103'5 93°5, 965, BS gfiG) 103'8) I0I"4 ret 95'2| rors 
' i 
in “British Rainfall, 1870.” I was so disappointed at 
the total disappearance of both the ten- and twelve-year 
periods, that I cannot say that I have closely scrutinised 
* Brit. Assoc. Report, 1865, p. 202. 
+ See Times, Nov. 12 and Dec. 3, 1872. 
