December 29, 1892] 



NA TURE 



21 



precipitation on trees, &c., which is increased by wind, and large 

 drops bf rain on to the ground beneath them. This condition 

 seems best explained by Aitl<en's discoveries showing the possi- 

 bility of a super-saturation of air when the numi er of dust- 

 particles is unusually small in a mass of air which is humid and 

 cooled to saturation. The dust-particles from their minuteness, 

 and from their inaliility to fall below the temperature of the air 

 owing to the cloud canopy above, do not condense much of 

 the vapour, and consequently any solid object of the same or 

 slightly lower temperature brings about precipitation from the 

 passing air, which may p issibly be super-saturated. A slight 

 fall of temperature in ihe air, or sometime-^ an increase of dust- 

 particles, then pro luces fog. A dry fog may thus result from 

 cold causing con lensation o.i a very large number of dust-par- 

 ticles which are radiating heat rather freely, and a damp mist 

 from partial condensation from super-saturated air on a com- 

 paratively small number of dust particles not radiating freely 

 owing to a clouded sky. 



These considerations explain why a dry fog is densest in 

 London and a wet mist den-^est in the country. A dry fog is 

 the work of cold radiating pirticles, a wet mist is the work of 

 cold air mixing with warm. " In a fog," says Angus Rankin,^ 

 " the watery vapour in condensing has more particles to condense 

 on, and consequently the particles of fog are smaller, and on 

 meeting with an object wiih a higher temperature, instead of 

 wetting it. the object dries them up by parting with some of its 

 heat. On the other hand, in a mist, the particles of dusr, being 

 few, have more water condensed on each, and so are larger and 

 do not readily evaporate with small increments of heat." Yet 

 in a damp mist the addition of a large number of dust-particles, 

 as in a town by day, scarcely increases the density of the mist. 

 In fact, the wet mist is less dense in London than in the country, 

 owing to the higher temperature and lower humidity of the air. 

 Dry or radiation f)gs, which cling to the ground, are the most 

 dense in smoky places. 



In fogs with frost in winter, such as have occurred several 

 times in the last few years, I have always found the windward 

 side of objects to be much more heavily frosted than the lee- 

 ward, and the rime to attach itself most to points and edges. 

 Trees have thus tiecome laden with rime, even so as to break 

 down branches ; iron points of railings, splinters of wood, wires, 

 and blades of grass have borne spikes and fern-like growths an 

 inch or more long, and heather and fern in hollows have been 

 whitened as if with a fall of snow. In weather of this kind it is 

 difficult to say what is dew or frost proper, and what is deposited 

 moisture from super-saturated air and from fog. On the same 

 night a white frost may present the characteristics of fog- 

 deposition in a valley and of clear condensation on a neighbour- 

 ing hill. 



Dew and frost are in fine the result of many causes which 

 inter-operate in a complex manner. The importance of the 

 laws of^ gases of the multitude of fine adaptations in the relations 

 of vapour, air, water, earth, and plants ; the importance, too, 

 of the thermal receptivity of boundless space, gives an mterest to 

 this branch of meteorology which is second only to its beauty. 



ARBORESCENT FROST PATTERNS. 



pROF. MELDOLA'S account of Arborescent Frost Patterns 

 -^ has excited a good deal of interest, and we have received 

 many letters on the subject, some of which we have already 

 published. To-day we give reproductions of photoijraphs we 

 have received from Mr. J. Maclear, Cranleigh. Fig. i re- 

 presents a photograph of a facsimile tracing of a '• Nature 

 print "of an ice crystallite taken by Mr. A. Anderson on a 

 still and sunny early morning in lanuary 1887, afier a not very 

 severe frost. The sunshine had just dried thf rest of the frost 

 off the flagstone, and left this mud and ice-ciystallizaiion, whi.h 

 he promptly secured on soft piper by means of a soft pad- 

 pressure, and I bus got a perfec Nature primed impression. 

 The original (now unforiunaiciy lost) showed an appearance 

 of vegetable (moss) growth, even more .strikingly than in this 

 tracing from it. 



With regard to Fig. 2 Mr. Maclear writes: — "The melting 

 ice under the dabbing pad formed a natural pigment with the 



^ Journal of the Scottish Meteorological Socieiy. Third Series. No. viii 



NO 1209, VOL. 47] 



mud on the flagstone, the rest of the flagstones being perfectly 

 dry already by the early morning sunshine." 



Prof. Meldola sends us the following interesting letter which 

 he has received from Corbndge on-Tyne : — 



"I was much interested by your note in Natitrk the other 

 day, anent the frost markings of a vegetable pattern. I have 

 seen just the same forms several times in the north, but it is I 

 think the least common of the patterns usually met with. I 

 write, however, to call your attention to Figs, i and 7 of , Plate 



Fig, I.— Ice crystallite, '• Nature printed " by A Anderson, January 1887. 

 Facsimile tracing by J. Maclear, January 1887. Size of crystal 14} 

 inches X 13J inches. 



vii, illustrating the article on Meteorology in the " Encyclopaedia 

 Metropolitana" (1845, vol. i. of plates, vol. v. of text). These 

 figures are very like yours and some of the others given with 

 them are also very interesting. I have often shown my students 

 when out in the fields in cold weather how exactly the mud- 



FiG 2.— Photograph from the original " Nature -print /' made M"",^ 

 Anderson, of an ice crystallite, January 1887. Size of crystal, 10} inches 

 X 7 inches. 



cum-frost markings of the common feathery volute type imitate 

 the so-called Caudagalli fossil fucoid (?) which is one of the 

 most abunHant objects on the surface of the cnrboniferous lime- 

 stone courses about here. As far as form goes they are identical, 

 and there is no structure to be discovered in the fossil markings. 

 " Corbridge-on-Tyne, December 16. G. A. LebouR. 



Prof. Sollas writes to us :— The correspondence on this 



ubject that has lately appeared in your columns (particularly 

 Prof. Bonney's reference), leads me to anticipate a communica- 

 •ion I horie shortly to present to the Roval Dublin Society on 



he growth of crystals. The arborescent forn.s assuined by ice 

 are merely a special case of a very general problem— that of the 



