NoVEMBKlt 1, 18!i3.J 



KNOWLEDGE 



209 



the air. Where the surface of the earth is warmed the 

 air rises, and the winds rush in from all sides. I3ut 

 though it is possible to conceive of uuequal heating of the 

 rare gaseous matter in the region just below the plioto- 

 sphere, where the photospheric clouds might be conceived 

 of as shielding certain portions of gas from radiation into 

 space, and in the region just above the photosphere where 

 the clouds may shield certain portions from the radiation 

 of the solar nucleus, it is much more difficult to conceive 

 of such unequal cooling in the region far below the photo- 

 spheric cloud layer, or of velocities being generated by slight 

 differences of weight sufficient to carry the rising matter far 

 above the level of the solar atmosphere. — A. C. Eanyaro.] 



THE COMPARISON OF PHOTOailAPHS. 

 To the Editor of Knowledge. 

 Dear Sir, — The method proposed by Mr. F. H. Glew 

 in Knowledge for October also occurred to me some 

 months ago. It may interest him to know that I have 

 put it to a practical test. Mr. Downing kindly sent me 

 from the Nautical Almanac office the positions of some 

 asteroids, which would be in a suitable position near the 

 meridian in the month of August. I took a photograph 

 of the region round the planet Arethusa on the !i2nd, and 

 the following night took another photograph of the same 

 region. I then took a positive by contact printing off the 

 negative of the 2'2ad, and superposed it on the negative 

 taken on the 23rd. The black star imai/i-f: of the negative 

 then exactly fill up the clear white images in the positive, 

 and the star imai/f.i then completely vanish. The image 

 of the asteroid, being in different positions on the two 

 plates, will not blot itself out, and it can then be seen at 

 once on inspection. I have had a frame made to hold 

 the two photographs in superposition, and provided with 

 screws acting on slides, so that the one photograph can 

 be adjusted with the greatest precision over the other. 

 The frame slides into an optical lantern, and the image 

 of the photograph when thrown on a screen can be quickly 

 examined for a planet. Youis very truly, 



W. E. Wn.soN. 

 Daramona House, Street, Co. Westmeath. 

 October 9th, 1893. 



LUNAR CRATERS SEEN IN RELIEF. 

 To tlu Editor of Knowledge. 



Dear Sir, — I have repeatedly observed that when 

 photographs of lunar craters, like the beautiful ones we 

 have lately had iu Knowledge, are viewed so that the 

 light from a window or lamp shines upon them from the 

 same direction as that of the sunlight when the photo- 

 graphs were taken, the craters appear as craters ; but 

 if the photographs be turned round so that the light from 

 a window or a lamp comes from the opposite direction, 

 there is immediately an appearance of false relief and 

 the craters seem to be raised up like mounds. The 

 effect is curious, and seems to show that the same drawing 

 and the same arrangement of light and shade can repre- 

 sent indifferently a hollow or an elevation. What is the 

 explanation "? 1 have seen the san'.e effect in looking at 

 the moon with a telescope ; the craters seem suddenly to 

 stand up like mounds. 



Alderbury Vicarage, Salisbury. R. S. Hutchings. 



Oct. 5th, 1893. 



l^It is evident that suitable shading, as well as perspec- 

 tive drawing, influences our minds m the compound 

 mental process gone through when we translate a picture 

 drawn in two dimeusions into objects existing in three 

 dimensions. A little consideration will show that exactly 



similar outlines might correspond to a hollow or an 

 elevation — a crater or a hill ; but with a side illumination 

 the disposition of light and shade would be reversed in 

 passing from the one to the other. And it would be 

 equally reversed if the objects were illuminated from the 

 other side. Consequently, before wc can make up our 

 minds whether a drawing represents a bas-relief or an 

 intaglio, we involuntarily consider from which side the 

 light is falling ; and a mistake made as to the direction 

 may cause us to interpret the drawing or photograph of a 

 crater as representing a mound with similar outlines, or 

 rice rersi't. — A. C. R.l 



THE SPECTRA AND MOTIONS OV STARS. - 

 To the Editor of Knowledgk. 



Sir, — I had intended making some remarks on the letters 

 of Prof. Kapteyn and Mr. Boraston, which have appeared 

 in your columns, on the subject of the spectra and proper 

 motions of the stars and their distribution in space. I 

 have, however, but little to write, and as I hope the new 

 volume of the Draper Catalogue, dealing with the southern 

 stars, will shortly be available, it seems desirable to wait 

 for further data. Prof. Pickering's paper, recently read at 

 Chicago, is a very suggestive one. It confirms one result 

 which I had derived from an examination of the proper 

 motions of the stars, viz., that stars of the Orion type (B) 

 are not to be regarded as intermediate between the Sirian 

 and Solar stars, but that, on the contrary, the ordinary 

 Sirians (A) are to be regarded as intermediate between the 

 Orion stars and those of the Solar type. He does not, 

 however, place the Capellan stars (F) at the end of the list 

 as the dullest and consequently the nearest to us (on the 

 average) for their magnitude, as I hold them to be ; and, 

 considering the brightness of Canopus, as well as its small 

 proper motion and parallax (according to Dr. Gill), it is 

 rather startling to find it referred to this type. Perhaps 

 the reference may turn out to be a mistake, like that of 

 Eigel, which is now referred to the Orion type (B). I 

 find also that Prof. Pickering refers the sun to the Arcturian, 

 not the Capellan type. The Galactic type he regards as 

 A. I am rejoiced to see that he contemplates deducing 

 the apex of the sun's way from a separate examination of 

 stars with different types of spectra. If the Sirians of the 

 type A, or the Galactic stars, form a separate system, the 

 fact will probably be revealed by finding a different apex 

 for the sun's way from that derived from stars with other 

 spectra. But the proper motions of stars of type A, and 

 still more of type B, are usually very small, and their true 

 motions are therefore liable to be overlaid by errors of 

 observation and computation. Determinations based on 

 Solar stars will, at all events, be more reliable for some 

 time to come. Sincerely yours, 



Dublin, October 17th. W". H. S. Monck. 



THE TINTS OF THE LUNAR PLAINS. 



By A. C. Ranyard. 



THE full moon, which appears so silvery white as it 

 shines in the evening sky, reflects, according to 

 ZoUner, only about one-sixth of the light which 

 falls upon it, and when its dazzling brightness is 

 partly quenched by daylight it still appears white, 

 and might easily be mistiiken for a small white cloud. 

 But as seen beside a vast ball of snow similarly illuminated 

 by the sun's rays, the moon would appear a dull grey, for 

 fresh-fallen snow reflects only 0-783, or rather more than 

 three-quarters of the light which falls upon it ; while the 



