394 



NATURE 



\_Fcb. 26, 1880 



Table III. — Winter Cloud. 

 Greenwich. Oxford. Munich. Br< 



1S26 



1834 



1846 +0-12 



1857 +0-I2 



1S68 +0-17 



— -0-03 — +0'C>9 



-0-34 -0-53 -1-27 -202 

 4-0-46 — +°'4S +1*08 



Means +0-13 +C03 



1829 — — 

 1837 



1S4S 

 1S48 



185S +°'52 +0-13 



i860 ±0'0 +0-21 



1866 -0-25 -0'54 



187O -0-28 -004 



-04I — 



-0'05 

 + 0-25 

 + 0-52 



±o - o 



-0-25 



-0-28 



+ 0-17 

 -0-58 

 +0-47 

 +0-52 



-0-02 



+ °75 



+ 0-10 



-017 



+ 012 



-°'33 

 + no 

 + 1-48 

 - ro2 

 -049 



Means +0-03 



-0'o6 +0-05 +0-14 



-Number of Days on which Schwaieteas unable to 

 observe the Sun at Dessau. 



Years. 



1S26 

 27 

 28 

 29 

 3° 



31 



32 



33 



34 



35 

 36 

 37 



38 

 39 

 40 

 41 

 42 



92 

 *4 

 121 



126 

 96 



121 

 166 

 197 



^5 

 160 



I°3 



S2 

 5S 



Years. 



1843 



44 



45 

 46 

 47 

 4 S 

 49 

 5° 

 5i 

 52 

 53 

 54 

 55 

 56 

 57 

 58 

 59 



Days. 

 41 

 46 

 33 

 5i 



Days on which neither Prof. Wolf nor his Assistant could observe 

 the Sun at Zurich. 



Years. 

 i860 



61 

 62 

 63 

 64 

 65 

 66 

 67 

 68 



Hot wave 



Days. 

 J2 



8l 

 76 

 90 



74 

 69 

 67 

 66 

 92 



1869 



70 

 7i 

 72 

 73 

 74 

 75 

 76 



77 



Days 

 101 



So far as the preceding tables afford a basis for deduction, it 

 appears that with few exceptions (1) the annual amou it of cloud 

 is below the mean at the epochs of the crests of the heat-waves, 

 and above the same at those of the cold-waves ; (2) that the 

 relation is of the same kind, but more marked when the results 

 for the summer season alone are compared ; (3) that the results 

 for the winter show- in several cases a tendency to vary in the 

 opposite manner. 1 



I may remark that in general the dates of the crests of the hot 

 and cold waves, as given by Prof. Smyth, coincide with, and 

 include, the principal critical epoch-; of the clou 1 variation. 



Judging from the cloud observations alone, the most intense, 

 as well as most universal waves would seem to have been the 

 hot waves of 1S57 and 1868, and the cold wave of i860. 2 



1 As a further addition to the evidence just g'.ven. both in favour of the 

 secular variation and the contrary character of the two extreme seasons as 

 to cloud, Prof. Piarii Smyth tells me that the results of the cloud observa- 

 : Edinburgh for eighteen years show June and July. 1879 ( ine d a e o( 



nths throughout 

 the whole being 

 18 (a heat-wa\ e ', 



th.j 



cold-w 



1 the cloudie 



_ j), to have bes 



the period, but December, 1879, 'he clc 



excessively cloudy. On the other hand June and July, : 

 were the clearest ever known. 



3 It is somewhat remarkable that in Dr. Kiippen's great work on the tem- 

 peratures in different parts of the globe in connection with the sun-spot 



It would be premature to attempt to draw any definite con- 

 clusions from the results I have exhibited, but they rather tend, 

 I think, to dissipate the notion Prof. Smyth apparently entertains, 

 that there is any specific difference between the waves of heat and 

 those of cold. 



It would seem indeed as though both were partially dependent 

 upon watery vapour and its transformations, the heat wave being 

 iu part the effect of an excess of sunshine, and the cold wave of 

 an excels of cloud. 



Again, were the heat waves of more direct cosmical origin 

 than the cold waves, they should occur more universally and 

 more simultaneously in different parts of the world than the latter, 

 whereas the results of most investigations into this matter point 

 the other way. The epochs of maximum and minimum annual 

 temperature may be respectively nearly identical for as large 

 a district as that included by the stations employed above, but 

 they certainly differ to some considerable extent, though at the 

 same time in a regular and progressive manner, when the 

 observations are made to embrace an entire hemisphere. 



Thus, according to Koppen, the following are the dates of 

 maximum and minimum air-temperature in the tropics and 

 extra-tropics respectively : — 



1 830" I 



1836-4 



IS47-6 



1831-9 

 1837-8 



i85°"3 

 (tS6i-6) 



Piani Smyth's 



.. 1826™ 



I834-5 

 1846-4 

 1857-9 



Piatzi Smyth's 



epochs of crests d£ 



cold-waves. 



1829-6 



l837'3 

 I845 2 

 1848-0 

 18558 



iS6o"3 



Wolf's dates for 

 minimum sun-spots- 



18233 

 I833-9 

 1843-5 

 1856-0 

 1867-2 



sun-spots. 

 1820-9 

 1837-2 



iS6oi 



From the above table it is evident that both the heat and cold 

 waves are retarded in the extra-tropics behind those in the 

 tropics, the mean lag being as much as 2 "9 years in the case of 

 ti.e former and 22 years in that of the latter. There is no 

 reason, therefore, for supposing either of these phenomena in 

 the extra-tropics to be the direct effects of solar or cosmical influ- 

 ences ; but, on the contrary, there is much to favour the notion 

 that they are both equally the indirect consequences of the corre- 

 sponding elevations 1 and depressions of temperature in the 

 tropics. 



It will be noticed that while the crests of both the hot an! 

 cold waves given by Prof. Smyth agree in the majority of cases 

 with those given by Dr. Koppen for the extra-tropics and also 

 with the sun-spot epochs, there are one or two cold waves, 

 such as those of 1845, 1S55, and 1866, which appear completely 

 isolated from either of these latter, though I am not aware that 

 they are inferior to the rest in point of magnitude. That even 

 these waves are not of mere local occurrence, though their 

 prototypes do not appear in the tropics, is probable, from the 

 fact that similar ones have been noticed by Dr. F. G. Hahn to 

 occur at Leipzig in 1S45, '855, and 1S65, in the form of 

 sec mil iry maxima of cold corresponding to the secondary 

 maxima in the aurora. 



In the short cycle 1829-37 no secondary wave appears at 

 Leipzig just as at Edinburgh. 



Meanwhile, whatever causes be ultimately adduced to account 

 for the appearance of these periodical waves of heat and cold, it 

 is evident that they partially bear out the designation accorded 

 them by Prof. Smyth, of "sunshine cycles." 



February 3 E. Douglas Archibald 



period, the heat-wave of 1857, as deduced from air-temperatures, appears 

 only as a I >cal phenomenon in the extra-tropics. The other dates, 1825-8, 

 1834-2, 1846-4, and 18687, given by Dr Koppen for the n.ax.ma of the tem- 

 pera. ure of the extra-tropics are nearly identical with those deduced from 

 the earth-temperatures by Prof. Smyth. 



1 This epjch is given by Prof. S. A Hill, of Allahabad, in continuation of 

 Dr. Koppen's work, and is deduced from observations taken in India (see 

 "Variations of the Rainfall in Northern India." by S. A. Hill, Indian 

 Meteorological Memoirs, p. 193). Creat reliance cannot therefore be givei 

 to it, though at the same time it agrees 

 sw£-tropics, as given by Dr. Koppen. 



ell with the result for the 



