ya7i. 28, 1886] 



NATURE 



293 



Besides these there were a few other small clouds, white, atid 

 of unusual character and brightness, but at no time did I detect 

 in them any certain trace of colouring. 



Deconber 31. — At about 10 a.m,, and for some time after, I 

 noticed a small coloured cloud, above, but a little to the west of, 

 the sun, and at an altitude of about 30°. The lower edge was 

 blue, immediately above this a narrow band of green, the rest 

 of the cloud being faint, almost colourless, but occasionally 

 showing a slight yellowish tinge, and at the upper edge a touch 

 of red. 



At sunset there was the most wonderful display. At 

 4 p.m. in all parts of the sky were bands of light cirrus cloud, 

 flushed with crimson by the setting sun. In the west, higher 

 than the cirrus, and shining through them and in the intervals 

 between, were splendid iridescent clouds, set off .against a back- 

 ground of blue. I have never yet seen anything to equal the 

 glory of this sky. It is impossible to describe it, the colours 

 were so varied and their changes so rapid. I confined my at- 

 tention, therefore, to the two largest and most beautiful clouds, 

 and the following is a record of the colours and their variations, 

 so far as I was able to observe them. 



(l) One of these clouds is shown in Fig. i. It was situated 

 a few degrees south of west, about 20° above the horizon, and 

 was about 15° in length and 5° in greatest breadth. During the 

 ■whole time it seemed stationary, though changing slightly in 



Fig.2. 



form. It was crossed by beautiful bands of colour, separated 

 by fairly sharp lines. Sometimes, when the stripes of colour 

 were narrow, I have included them in one band, to save time in 

 drawing ; the order being then given from left to right. 



4.5p.m. — Fig. I, a. The right end of the cloud hazy. (i)Green, 

 (2) yellow, orange, red ; (3) blue, green, yellow, red ; (4) red, 

 with a tinge of purple ; (5) bright purple ; (6) red, green. 



4.11. — Fig. I, i5. Both ends hazy, (i) Pink in the haze, blue ; 

 (2) yellow, green, orange ; (3) green ; (4) red ; (5) blue ; (6) red, 

 ending in green in the haze. 



4.18. — Fig. I, c. The cloud rather longer and narrower than 

 before, (i) Green; {2) green; (3) yellow, orange ; (4) green ; 

 (5) red ; (6) green ; (7) red. 



4.21. — The orange band 3 beginning to invade the green 

 band 2. 



4.22.— The red band 5 growing brighter. 



4.25. — The red and orange bands, 3 and 5, widening, and 

 becoming the predominant colours of the cloud. 



4.27. — Fig. I, d. (i) Light haze; (2) hazy; (3) red; (4) 

 green; (5) bright red; (6) greenish; (7) orange red. The 

 general hue of the cloud was at this time reddish orange, the red 

 and orange bands being much brighter than the rest. 



4.30. — Fig. I, c. The outline hazy, (i) Haze; (2) red: (3) 

 reddish orange ; (4) red. 



4.32.-- Fig. I,/, (i) Thin haze; (2) not quite so bright as 

 belote, but all of a deep rose-colour. 



4- 34- — I'he rose-coloured part had a slight tinge of purple, 

 which, tWi) minutes later, had become more marked, but rising 

 clouds now stopped further observation. 



(2) The other cloud was in a west-south-west direction, about 

 10° above the horizon, and partly hidden by a bank, which 

 served, however, to show how slightly the cloud altered its 

 position. 



4.7 p.m.— Fig. 2, a. (i) Green ; (2) orange red; (3) light 

 green ; (4) violet. All these colours very bright. 



4- 13- — The colours had changed, so that the general hue of 

 the cloud was blue. 



4.14. — Fig. 2, b. (i) Orange; (2) bright blue; (3) violet. 

 The right edge hazy. 



4.16. — The left side bhte, the right violet, with a narrow pink 

 band on the lower edge, 



4- 17. — The greater part of the cloud a very bright light blue, 

 violet on the right, pink on the upper edge towards the left. 



4.22. — Very faint and bluish. 



4.25. — The cloud smaller, and bluish-green, but still faint. 



4.26. — Green, and brighter. 



4.29. — Fig. 2, c. The whole cloud much brighter, though 

 not so bright as when first seen. (l) Reddish; (2) green; 

 (3) reddish. 



4.34. — Ihe cloud the same shape as at 4.29, and the whole of 

 it orange-coloured. After this moment it was hidden by heavy 

 clouds. 



Beside the two clouds above described, and several othev 

 smaller ones similar to them, there was visible in the west, at 

 4.26, along narrow band of cloud (about 20° long and 3° or 4° 

 broad), parallel to the horizon, and of a distinctly violet colour 

 throughout. Ch-\rles D.-wison 



Sunderland, January 12 



Parallel Roads in Norway 



In Mr. Hansen's account of the terrace formation of Central 

 Norway he discards the sea theory of their origin, as well as the 

 detrital dam, the local glacier theory, and also that of Prof. 

 Prestwich, of landslips. The cause he ascribes to rests in the 

 passage of the inland ice seawards, allowing lakes to form in the 

 watershed while ice remained in the valleys seaward. Does this 

 idea not reverse the order of Nature ? Would it not be far 

 simpler, more reasonable, and more in accordance with the laws 

 of Nature to conclude that ice would remain in the highest 

 valleys of the country longest, and that the parallel roads or 

 terraces arc the ice margins or lateral moraines where the ice 

 rested after the most intense glaciation ceased, while the surplus 

 passed over the cols, and the passage seaward w.as more or less 

 retarded by the configuration of the country ? The Lochaber 

 roads are mostly composed of the usu.al glacial stuff of the dis- 

 trict, it is neither washed as lake margins or sea beaches. The 

 only water-washed material seems to have run down from the 

 hills above, before glaciation ceased, and vegetation covered the 

 surface. The roads are neither strictly parallel or horizontal, 

 and just what might be expected to be formed by ice lying for a 

 long time in a valley when the g'^owth did not greatly e.tceed 

 the waste and the motion w.as slow. J.\MES Melvin 



Edinburj^h 



Dew 



Haying read with interest the abstract in Nature ot 

 January 14 (p. 256) of Mr. Aitken's observations on dew, I 

 noted attentively during a walk this morning the behaviour of 

 the hoar-frost as deposited on different objects. The morning 

 was fine and frosty after a clear cold night. There was a 

 copious deposit of hoar-frost upon the grass, upon the upper side 

 of wooden rails, and upon the topmost twigs of the bushes in 

 the tall hedges (6 to 8 feet high), but the lower twigs in the 

 hedges had little or none. On stones in the road, as Mr. 

 Aitken observes, there was little hoar-frost on the upper surface, 

 only lines of ice crystals along the salient angles, but their under 

 surfaces were thickly covered. With the loose heaps of broken 

 stones by the road-sides the case was diff'erent : here the upper- 

 most stones were thickly coated with frost on their upper sur- 

 face, but had little on their lower surface ; the stones underneath 

 the uppermost layer, on the contrary, were coated with hoar- 

 frost on their under, but not on their upper, surfaces. The 

 hollow " cat's ice " on the road-side puddles, where previously 



