480 



NA '1 URE 



{March 26, 1S85 



at 8.30 it reached the zenith. In this position— from the 

 northern horizon to zenith— the phenomenon remained almost 

 uninterrupted all the following time. The light was 

 rather feeble, and in the beginning the motions were insigni- 

 ficant. Hut at 10 o'clock the peculiar blazing or undulating 

 movement that is designated by the name of coruscation, began 

 to be seen, and during four hours and a half at least the whole 

 northern half of the sky was the theatre of this uncommonly 

 violent chase of the luminous clouds. The culmination of the 

 aurora happened at 10.30, when on the northern sky advanced 

 a series of splendid streamers, the inferior points of which played 

 in red and green. This radiance was only of short duration, 

 and later there appeared in the north only arches more or less 

 distinct ; while on the higher parts of the heavens the cha-ing 

 flames incessantly continued their playing. Still, so late as 

 14.30 saw the flames as far as to the zenith with unimpaired 

 violence. 



I may add that on this occasion I succeeded in what I myself, 

 as well as other friends of the aurora, have tried before in vain, 

 viz. to get the aurora to make impression on a photographic plate. 

 I exposed in all five platen ; of these four (for times of exposure 

 of 2-4 minutes) without the least trace of action, but the fifth, 

 which was exposed during 8J minutes, shows both a part of the 

 horizon with a high church spire and a feeble representation of a 

 portion of the aurora. I must, however, state that this portion 

 in itself was but very feebly illuminated, and that at the time when 

 the phenomenon developed the greatest power of light I was 

 prevented from applying the camera. 



The object-glass employed was : Voigtlandereuroscope, No. 1 ; 

 Schleussner's dry plates. 



Also on the 16th, in the evening, 8.45 to 10, there appeared 

 an aurora, but consisting essentially only of feeble fragments 

 of arches rather low on the northern sky. The aurora has in 

 recent times been astonishingly rare : here in Christiania, in the 

 course of the whole winter, it has been observed on the follow- 

 ing days :— September 14, 17, 24; October 14, 15 ; November 

 17 ; December 22 ; January 22 ; February 14, 16 ; March 12, 

 15, 16. 



March 15 excepted, all these aurora: have been rather 

 insignificant. Sophus TROMHOLT 



Christiania, March 17 



"Peculiar Ice Forms" 



Under the above caption, several correspondents of 

 Nature have recently described and discussed the agglutin- 

 ated filamentoid ice-crystals commonly extruded from un- 

 frozen earth under suitable conditions of moisture and tem- 

 perature. B Woodd Smith records their occurrence in the 

 Savoyan Alps (his language implying variety of the phenomena 

 there), and attributes them to the linear expansion incident to 

 congelation of capillary columns of water in a thin sheet of 

 soirresting upon rock (vol. xxxi. pp. 5-6). W. alludes to 

 such crystals in general terms, refers to a previous notice of 

 similar phenomena, and (erroneously) allies them genetically to 

 hoar-frost {id., p. 29). Dr. John Rae discusses distinct (but erro- 

 neously supposed similar) phenomena at length, and argues that 

 the several strata of crystals are remnants of successive sheets of 

 ice or snow (id., pp. 81-2). Mr. Smith then controverts 

 Mr. Rae's explanation, maintains his own, and refers to several 

 earlier communications in Nature relating to filamentoid ice- 

 crystals (27'., pp. 193-4) ; and sebsequently he transmits a letter 

 from John D. Paul, who has essentially repeated his own obser- 

 vations in the Alps (:/>., p. 264). 



Now that it has become fashionable to revive forgotten 

 records, it may be pointed out that these correspondents ignore 

 the more valuable portion of the literature of their subject. Even 

 in Nature discussion of the fibrous ice-crystals extruded from 

 moist earth, wet wood, &c, was epidemic fifteen years ago, and 

 again ten years later, besides the -poradic cases of three years 

 ago, as shown in the following bibliography : — 



1870 Vol. I. — C. Spence Bate p. 55° 



,, ,, —Mr. Pengelly ]■■ <»-7 



,, Vol.11. — T. G. Bonney, John Langters (sic) pp. 141-2 



1S71 Vol. III. — T. G. Bonney, John Langton pp. 105-7 



,, ,, — T. G. Bonney p. 2S8 



1880 Vol. XXI. —Argyll p. 274 



,, — K. Meldola p. 3° 2 



,, ,, —Argyll P- 3 oS 



„ „ — O. Fisher P- 39° 



1880 Vol. XXI.— D. Wetterhan ... 

 ,, — L. Bleekrode 



„ ,, — S. T. Barrett 



,, — Win. Le Roy Broun 



',', Vol. XXII.— John Le Conte ... 



— R. H 



1882 Vol. XXV.— T. F. Duthie 

 „ Vol. XXVI.— H. Warth 



... p. 396 

 • •• P- 444 

 ... P- 537 

 ... P- 589 

 ... p. 54 

 ... pp. 145-6 

 P . 78 

 . ... p. 81 



The second outbreak was practically terminated by the com- 

 munications of Profs. Broun and Le Conte. The first of these 

 gentlemen wrote from a locality in which the phenomena are 

 readily observable, while the latter called attention to his own 

 elaborate researches of thirty years before (Proc. Am. Assn. 

 Adv. Sci., iii. 1850, p. 20-34; phiL Afa g-> third senes ' xxxvl 

 1S50, pp. 329-42). 



More recently (February 6, 1884) Prof. Schwalbe has placed 

 before the Physical Society of Berlin the results of his observa- 

 tions upon filamentoid ice-crystals in the Harz. After thorough 

 study he accepts Le Conte's views as to their genesis. 



Prof. Le Conte's explanation (which is essentially identical 

 with that subsequently offered independently by Prof. Broun) is 

 as follows :—" Let us suppose a portion of tolerably compact 

 porous and warm earth saturated with moisture, to be exposed 

 to the influence of a cold-producing cause. The soil being an 

 impeifect conductor of heat, only a very superficial stratum would 

 be reduced to the freezing point. As the resistance to lateral 

 expansion is less at the surface than it is at] a sensible depth 

 below, the effect of the first freezing would be to render the 

 apices of the capillary tubes or pores conical or pyramidal. The 

 sudden congelation of the water, filling the conical capillaries 

 in the superior stratum, would produce a rapid and forcible ex- 

 pansion, which, being resisted by the unyielding walls of the 

 cone, would not only protrude, but project or detach and throw 

 cut the thread-like columns of ice, in the direction of least resist- 

 ance, or perpendicular to the surface. This would leave the 

 summits of the tubes partially empty— a condition essential 

 to the development of capillary force. Under these circumstances 

 capillary attraction would draw up warm water from beneath, 

 which, undergoing congelation, would, in like manner, elevate 

 the column of ice still higher ; and thus the process would go on 

 as long as the cold continued to operate on unobstructed capil- 

 laries, supplied with sufficient water from below. It will be 

 remarked that this explanation makes the whole process of pro- 

 trusion to take place in a stratum of earth of almost inappreciable 

 thickness. It also presumes that the protruding force act[s] 

 paroxysmally" (Proc. Am. Assn. Adv. Sci., op. cit., 30-31). 

 He subsequently remarks (Nature, op. cit.) : "... In rela- 

 tion to the explanation of the phenomena, I have nothing to 

 add to that given" above, "except in relation to two points, 

 viz. : (1) that I did not sufficiently emphasise the importance of 

 the fact that the water contained in the capillary tubes in the 

 upper stratum of earth is cooled many degrees below the freezing 

 temperature ; and (2) that consequently the congelation would 

 necessarily take place paroxysmally." 



It may be pointed out also that the great majority of the cor- 

 respondents, both recent and earlier, base their explanations 

 upon isolated observations of phenomena rare in their localities. 

 In reality the extrusion of filamentoid ice-crystals is even more 

 common in certain localities than is indicated by Le Conte s 

 papers and Broun's letter. Thus, in the cultivated fields of the 

 Mississippi valley, during a cloudy day following an autumnal 

 rain, with an air temperature just below freezing-point, the 

 writer has seen a thin layer of soil elevated from one to three 

 inches over fully four-fifths of the area visible from the road 

 throughout a day's journey. Greater length is sometimes 

 attained by the ci -ystals. Within a week the writer has observed, 

 alon<r the roadsides just beyond the limits of Washington, many 

 irregular patches and belts of straight or slightly curved fila- 

 mentoid crystals, four, six, and even eight inches in height. 

 They were sometimes highest where they supported the greatest 

 weight of earth, leaves, twigs, or pebbles upon their summits. 

 In one case a worn quartzite pebble 1X1SX2I inches was 

 hoisted on a slender tuft of icy needles six or seven inches long, 

 fully two inches above the smaller neighbouring pebbles and the 

 film of soil in which it had been imbedded. 



While Le Conte's theory of the formation of the filamentoid 

 crystals extruded from moist earth or wet vegetal stems is accept- 

 able in a general way, repeated observation upon crystals 

 apparently in process of development has convinced the writer 

 that their growth is not paroxysmal. The effect of capillarity 



