October 22, 1891] 



NA TURE 



59i 



the vapour-laden air inside or near the fumarole, might be the 

 cause of a rapid and continuous condensing of the invisible 

 vapour. I noticed that the " powdering " of the air with any 

 kind of dust increased the cloudy column issuing from the 

 Bocca of the Solfatara. I am therefore led to believe that the 

 action of a paper- or faggot-flame in causing the increase of 

 visible vapour from the Bocca of the Solfatara is due both to the 

 production of carbon dioxide and to the increase of sqlid 

 particles of soot and of light unbumt fra^jments made to rise 

 and float in the air. 



These experiments may help in explaining the action of ex- 

 plosives in causing a downfall of rain. Not only does the 

 explosion produce a certain amount of carbon dioxide, but dust 

 is widely scattered in the air, and carried upwards by the hot 

 gases produced in the explosion. If the results of the experi- 

 ments in Texas and Kansas by General Dyrenfurth and Prof. 

 Curtis be confirmed, it would be interesting to see if the con- 

 densation of vapour in the atmosphere could be better insured 

 by purposely increasing the quantity of dust produced in each 

 explosion. The eflect would perhaps be enhanced if the dust 

 were of a markedly hygroscopical nature : the scattering in 

 high air of very minute particles of calcium chloride should 

 help in the making of cloud and rain. Italo Giglioli. 



Laboratory of Agricultural Chemistry, 



Royal Agricultural College, Portici, near Naples, 

 October 12. 



Weather Cycles and Severe Winters. 



The following view of the relations of severe winters is one 

 which I do not remember to have seen stated. 



Consider the 79 years 1812-90 (at Greenwich), and let us 

 take, as a measure of winter cold, the mean temperature of the 

 three months December, January, and February. Divide the 

 series of years at i860 ; giving a first series of 48 years (1812- 

 59), and a second of 31 years (1860-90). 



Now consider the first series. The coldest winter in it is 

 1813 (meaning, by that, 1813-14). The coldest of the following 

 winters is 1829 ; the coldest of the following, 1840 ; then come 

 (reckoning similarly) 1844 and 1846 (equal), 1854, 1859. The 

 absolute order of decreasing severity is to some extent the same, 

 but at certain points the order of time is reversed. 



Next take the second series. The coldest winter in this is 

 1890 {i.e. 1890-91) ; the coldest of those preceding, 1878 ; the 

 coldest of those preceding, 1870 ; then come (similarly) 1864 

 and i860. 



Thus we have a succession of severe winters of decreasing 

 severity, and another, after it, of growing severity. 



We may tabulate the data : — 



Mean 

 temperature. 



37-4 



37-1 

 36-4 

 34-6 

 341 



These data, put into the form of a graphic diagram, give 

 a wave whose crest (mildest of the severe winters) we seem to 

 have passed in the sixties. And it would appear judging by the 

 past, that we have not yet reached the bottom of the hollow ; 

 but that after some years' interval we may have a winter even 

 more severe than last, possibly we may have more than one, of 

 growing severity. 



It is right to stale that, as far as 1856, the values of mean 

 temperature used are those of Mr. Belleville, reduced to sea-level, 

 as given in a paper by Mr. Eaton to the Royal Meteorological 

 Society (Quarterly Journal, January 18S8) ; after .that date, 

 those of Greenwich Observatory, published annually. The slight 

 difference in kind does not materially affect the result. 



In the Meteorologis;he Zeitsckrift for September, M Woeikof 

 considers the question whether winters in Ru-;sia have been 

 growing warmer, and hi- examination of the St. Petersburg 

 records, from 1744 to 1890 (noting the number of cold days), 

 leads to an aflirmative answer. The number of very cold days 

 has, on the whole, fallen off considerably in the later sixty-three 



NO. I 147, VOL. 4.4] 



years compared with the earlier, and in the second half of our 

 century, as compared with the eighteenth and the earlier half 

 of the nineteenth. 



This, he finds, corresponds with popular opinion for Northern 

 and Central Russia, according to which intense frosts have become 

 more rare ; but in the south, in the Crimea, the Caucasus, and 

 Turkestan, there have been complaints of colder winters of late. 



Mr. Glaisher some time ago expressed the view that our winters 

 had been becoming milder. I have seen a criticism of this view, 

 to the effect that the proximity of Greenwich to such a rapidly 

 growing city as London might have to do with such a result. If 

 the facts are as I have suggested above, a growing severity has 

 taken the place of growing mildness, and the criticism referred 

 to would fail to apply. A. B. M. 



A Lunar Rainbow. 



On the evening of Saturday, October 17, at about 6.30 p.m., 

 the rare and interesting phenomenon of a lunar rainbow was 

 observed from Patterdale, Westmoreland. On the south-east, 

 the moon, which had just risen, brightened the sky behind the 

 mountains, while on the north-west there hung a uniformly dark 

 and unbroken screen of haze or rain-cloud, which lightened 

 off somewhat and was more scattered on the extreme west. 

 With its highest point lying almost exactly north-west, a semi- 

 circle of pale whitish light was projected against this vapoury 

 curtain. The bow was quite complete, but much brighter and 

 sharper on its northern arc than on that falling south. The 

 brighter portion fell over weird and clear into Glenridding (a 

 favourite haunt of sun-painted rainbows), and as seen striped 

 against the dark hill-sides of that valley, appeared to emit a 

 pale blue phosphorescent glare. At one time a shred of the 

 dark smoky haze scudded over, but did not completely obscure 

 the highest reaches of the spectral light. The radius appeared 

 smaller than in the case of an ordinary solar rainbow, and the 

 breadth of beam was about one- half thereof, or perhaps rather 

 less. The spectacle having lasted for about eight minutes, 

 light rain began to fall, and then the sky in a very short time 

 became quite clear and star-lit, and all was over. 



P. Q. Keegan. 



Patterdale, Westmoreland, October 17. 



The Destruction of Mosquitoes. 



The recent mention of this subject in your pages reminds me 

 that I was told a few years ago by an English gentleman who has 

 a most beautiful place on the Riviera that he had freed his 

 property from this pest. 



The property in question is a peninsula, and for that reason 

 is exceptionally open to separate treatment. On the Riviera, as 

 mai.y of your readers will know, fresh water is a somewhat rare 

 commoditv, and all of it that the inhabitants can lay hold of is 

 stored for future use in tanks or small receptacles. 



The larva of the mosquito lives, as I understand, only in fresh 

 water. Consequently, on the Riviera he is found in the tanks I 

 have named. 



The carp is, I am told, passionately fond of the larva of the 

 mosquito, and the Englishman I refer to had extirpated the 

 insect by putting a pair of the fish in every lank. 



The plan is not one that could be adopted everywhere, but it 

 is worth bringing under the notice of those whose circumstances 

 are like those of the Riviera. S. A. M. 



Law of Tensions. 



Possibly many science teachers find some little difficulty in 

 satisfactorily demonstrating to a class the " law of tensions " 

 for vibrating strings. In practice, unless the sonometer is 

 fixed vertically, the error introduced by friction at the pulley 

 (especially with heavy weights) is so great that the real tension 

 is very different from that represented by the weight attached. 

 Even if the apparatus be thus fixed, the changing of the weights 

 occupies time, and a comparison wire is necessary, which must 

 first be tuned to exact unison. The following admirable and 

 very simple method was suggested to me by one of my students, 

 and possibly there are some teachers to whom the idea is new. 



Instead of applying ten.^on by attaching weights, the result 

 may be effected much more readily by means of an ordinary 

 spring suspension-balance, such as is often used for weighing 



