560 



NA TURE 



{Oct. 9, 1884 



the arrival of the unusual sunsets last year, and that it has never 

 been seen here before, at any rate within the last twenty years. 

 This impression is founded, first, on the fact that, like Mr. Back- 

 house, I have been in the habit of frequently looking at solar 

 ha'os for years past. Secondly, I have been engaged since 

 September 8 last year in a series of observations with anemo- 

 meters attached to a kite-string (latterly wire), which has natu- 

 rally necessitated my frequently looking up at the sky. I 

 remember noticing the halo in November', and calling the atten- 

 tion of my assistant to the beautiful salmon colour it showed in 

 the interstices of a mackerel sky, which shut off the direct glare 

 of the sun. 



On several occasions I measured its radius with a theodolite 

 I was i^ing, and in every case the value came out either 22i° or 

 23 . It must therefore be due, like the ordinary ice-halo, to 

 refraction through hexagonal prisms. I noticed it all the way 

 going out to America in August last, and saw it beautifully 

 when standing on the Terrapin Bridge over Niagara Falls on 

 August 23. A large black cloud had shut off the sun's glare, 

 and the red border happened just to coincide with the edge of 

 the cloud. I called the attention of some American ladies to 

 the spectacle, which they had of course never noticed before, 

 and which they admired immensely. I regard the sunset 

 glows as partly an intensification of the halo, produced by the 

 greater thickness of the stratum through which the refraction 

 takes place when the sun is below the horizon (the sunset, in 

 fact, being formed of the upper half of the halo which lingers 

 behind) plus reflection from the same stratum, which can of 

 course only come into operation when the sun shines obliquely 

 on it. The fact of the halo remaining constant while the sunset 

 glows have become weaker may be readily explained on the 

 hypothesis that the stratum has gradually sunk to a lower level 

 than it had last winter, since the duration and even brilliancy of 

 the glow must of course vary directly with its height, whereas 

 the halo at midday need not be sensibly altered by" a variation 

 in its level. 



Three weeks ago yesterday, I saw the sun rise from the summit 

 of Mount Washington, 6293 feet above sea-level, and at ten 

 minutes to five o'clock saw, in addition to the usual sunrise 

 effects, a large circle of rosy purple haze situated about 15° above 

 the horizon, and apparently having no connection with the yellow 

 and red horizontal bands beneath it. The morning was ex- 

 quisitely fine, the only clouds being a few light streaky cirro- 

 strati, as shown in diagram, and the air was as clear as it only 

 can be in America. I may add that the haze circle appeared 

 almost suddenly after the first yellowish light had been visible 

 for twenty minutes, and as soon as the sun rose above the 

 horizon it seemed to vanish almost entirely. I have seen the 

 sun rise from various altitudes up to 12,000 feet, but I never saw 

 anything so curious as this sunrise before. In fact, I went up 

 mainly to see it, and was not disappointed. 



E. Douglas Archibald 



Punbridge Wells, September 27 



Cole's Pits 



Your note on the result of Gen. Pitt-Rivers' examination of 

 the Pen Pits (Somerset) in this week's number of Nature 

 (p. 545) reminds me of a series of similar pits in this county of 

 Berks, known as "Cole's Pits." They are situated near the 

 branch of the Great Western Railway which runs from Upping- 

 ton to Faringdon. I have visited them more than once when on 

 visits to Wadley, the residence of my friend Mr. T. L. Goodlake, 

 J. P. They correspond generally with the description given in 

 Nature of the Pen Pits, are probably of quite equal extent, are 

 similarly situated on rising ground forming the cap of a ridge of 

 hills, and are on the same Greensand formation. Many and 

 various theories have been propounded from time to time by 

 antiquarians, and of course the name by which these pits have 

 been known for generations has been appealed to as connecting 

 them with the "merry King Cole," and giving support to the 

 views of those who regard them as traces of an " ancient 

 British town." The utter absence of order in the arrange- 

 ment of these shallow holes and rude mounds (for they arc- 

 nothing more) excited my suspicion, nor could I see much 

 in them to suggest occupation by any race which has inhabited 

 these islands even so late as the time of the ancient Britons. On 

 further investigation of them I came across a more modern sirt 

 of hut, consisting of a space rudely roofed over, the back of 



which was cut into the side of one of these grass-grown banks. 

 The idea occurred to me that this would help to furnish evidence, 

 since, if these hollows were dug in the strata of the hill to be 

 roofed over for human habitation (the notion which, I am 

 told, finds general favour) we ought to find some traces of strati- 

 fication in a section thus presented to us. Not a trace of this 

 was to be found ; the section showed nothing but a chaotic 

 mass of rubbly material with no more order in its arrangement 

 than is to be found in the waste heaps of any old quarry or in a 

 terminal moraine. 



So far one's first impressions of the " pits" and mounds, as of 

 an extensive series of old disused quarries, received confirmation. 

 I cannot hazard a conjecture as to the extent to which quern- 

 stones may have been obtained from the "Cole's Pits" ; but it 

 seemed to me extremely likely that the range of hills in question, 

 here so many hills on sandy formations (our Bagshot Sands, for 

 example) owed its existence, qua hills, to the protection of the 

 hard " paw" which is so frequently met with in such formations, 

 which I have attempted to account for in the Proceedings of the 

 Geological Association (vol. viii. No. 3), and which was certainly 

 largely worked in places for rude building construction in very 

 early times, large masses of it being seen still in the old Roman 

 wall which to-day completely encircles the area occupied by the 

 ancient town of Silchester. It is possible, too, that such iron- 

 stone may have been rich enough in the metal to serve as ore, 

 when iron "forges" were common in past centuries in the 

 forest districts of the south of England. Further support is given 

 to the view which I venture here to put forward as to the origin 

 of the Cole's Pits, by the fact that lower down the hill, and at a 

 rather lower "horizon," there is a sand-pit open in the side of 

 the hill, in which thin bands of ironstone occur, some of which, 

 to judge from their specific gravity, might certainly have been 

 available as ore, under conditions which obtained in the iron 

 industries of the country a few centuries ago. A. Irving 



Wellington College, October 3 



The Flow of Streams 



The observations made by Mr. Maw on a stream flowing into 

 the Lake of Thun are an extremely interesting example of well- 

 known hydraulic laws. It would be well, however, if he would 

 tell us what was the depth of the water ; from the observed 

 phenomena I presume it was small. Mr. Smith's very ingenious 

 mode of practically showing the different strengths of a current 

 from surface to bottom should also be supplemented by an 

 account of the size of the stream operated upon. The situation 

 of the point of greatest velocity varies considerably with the con- 

 ditions of the river or stream, and is by no means fixed. The 

 observations made on some large rivers, notably those by Mr. 

 Revy on the large rivers of South America, seem to show that 

 the greatest velocity is, in such cases, almost, if not quite, on the 

 surface, and that from that point it diminishes uniformly down- 

 wards to the bottom. The ratio between the surface and bottom 

 velocities is, however, a constantly changing one, and in large 

 rivers varies with the depth of the water. Mr. Revy's observa- 

 tions seem to show that in large rivers, where the influence of 

 the banks is practically inappreciable, the surface velocity varies 

 directly as the depth, whilst the bottom velocity varies as the 

 square of the depth ; as the depth increases, the difference between 

 the top and bottom currents diminishes, until at a depth of about 

 71 feet they are practically equal. These results are borne out 

 by the observations of Messrs. Humphreys and Abbott on the 

 Mississippi and Mr. Gordon on the Irrawaddy. All observations 

 to be useful, therefore, should give the size of the river or stream 

 operated on. If Mr. Smith has not seen the diagrams given by 

 Mr. Revy in his " Hydraulics of Great Rivers," I think he would 

 tie interested in them. George Higgin 



Lepidoptera 



I AM at present carrying on some researches here, for which 

 a supply of living caterpillars of such large Lepidoptera as the 

 Death's Head, Goat, or Hawk Moth is necessary. As I have 

 had some difficulty in obtaining these, I should feel exceedingly 

 obliged to any of your readers who may find or possess larvae 

 of the above or other large species, if they could furnish me 

 with specimens. G. Lovell Gui.land 



Zoological Laboratory, University College, 

 ( lower Street, London, October 4 



