52 



NA TURE 



\Nov. 19, 1874 



being always a point at the same ilistancc on the other 

 side of the plane of reference, and the horizontal distance 

 between the two bcinj; the same as in the preceding case. 

 This property is illustrated by Fig. 2. 



It is obvious that the conjugate foci will occur not in 

 pairs merely, but in sets of unlimited number ; that is to 

 say, raised proceeding originally from any one point will 

 converge in succession to an indefinite number of other 

 points, which will be alternately on opposide sides of the 

 plane of reference. As every point on the surface of an 

 object will thus have its conjugates, we shall have a suc- 

 cession of images of the object. The first image will be 

 upside down, the second erect, and so on alternately. 

 They will be what are technically called " real " images, 

 and will be precisely equal and similar (except as regards 

 inversion) to the object itself. It is of course to be un- 

 derstood that the action here described is confined to one 

 dimension only, resembling that of a cylindrical rather 

 than of a spherical lens. Rays are bent to and from the 

 plane of reference, but in no other direction. This theo- 

 retically simple case is so important foi the light which 

 it throws upon the possibilities of atmospheric refraction, 

 that we shall examine some of its consequences a little 

 further. 



What will be the appearance presented to the eye of 

 an observer in any given position ? 



The case differs greatly from that of the images in ordi- 

 nary optics, where the refracting instruments are glass 

 lenses, and the eye sees the image by means of rays 

 which travel in stiaight lines. 



In the case now before us, the observer will in general 

 see a virtual image, differing considerably, both in size 

 and direction, not only from the object itself, but also 

 from any one of the real images. The apparent direction 

 of any point of the visible image is of course determined 

 by drawing a tangent to the ray which enters the eye* 

 (Figs. 6 and 7) ; and the visual angle, or, as we may call it, 

 the apparent size of the object, will be ihe angle between 

 two of these tangents. If the eye is a little distance (say 

 a few feet) behind one of the real images, enormous mag- 

 nification will be produced, for the image has the same 

 linear height as the object, and is seen from a distance of 

 a few feet, instead of from the real distance of the object, 

 which we may suppose to be a few miles. We shall thus 

 have enormous magnification of the vertical diameter of 

 the object, while the horizontal diameter will of course be 

 only of the natural size, since the rays have undergone 

 no bending except up and down. An object whose 

 breadth is equal to its height will thus be magnified into 

 a tall column. Some appearances of this kind, copied 

 from Scoresby's " Greenland," are represented in the first 

 two figures of Plate II. The following is Scoresby's de- 

 scription C Greenland," p. 96) : — 



" Hummocks of ice assumed the forms of castles, 

 obelisks, and spires, and the land presented extraor- 

 dinary features. In some places the distant ice was so 

 extremely irregular, and appeared so full of pinnacles, 

 that it resembled a forest of naked trees ; in others it 

 had the character of an extensive city crowded with 

 churches, castles, and public edifices." 



Again, on page 163 of the same work : — 



" At one period the phenomenon was so universal that 

 the space in which the ship navigated seemed to be one 

 vast circular area, bounded by a mural precipice of great 

 elevation, of basaltic ice." 



The magnificent columns which constitute a portion of 

 the wonders of the Fata Morgana, at the Straits of Mes- 

 sina, are in like manner to be attributed to vertical mag- 

 nification. And an appearance of the same kind, known 

 as "the merry dancers," is often seen by boatmen oif the 

 Giant's Causeway, in looking over the Skerries towards 

 Portrush. 



• Tlie letter E, in all the figures, denotes the position of the observer's 



ll we could have density distributed symmetrically 

 round an axis, instead of on the two sides of a plane, we 

 might of course have magnification without distortion. 

 Rut we can scarcely conceive of any arrangement at all 

 resembling this existing in the atmosphere. 



It is further to be remarked, that the apparent distance 

 of one of our colunm.ar images from the observer's eye is 

 an ambiguous quantity. If judged by left and right dis- 

 placement, it is the real distance of the object, If judged 

 by up and down displacement, it is much less, being 

 approximately the distance of the real image. 

 ( To be continued.') 



SOME REMARKS ON DALTON'S FIRST 

 TABLE OF ATOMIC WEIGHTS* 



A S the Society is aware, the first table, containing the 

 -'*- relative weights of the ultimate particles of gaseous 

 and other bodies, was published as the eighth and last 

 paragraph to a paper by Dalton on the absorption of 

 gases by w.iter and other liquids, read before this Society 

 on Oct. 21, 1S03, but not printed until the year 1S05. 

 There appears rc.ison to believe that these numbers were 

 obtained bv Dalton after the date at which the paper was 

 read, and that the paragraph in question was inserted at 

 the time the paper was printed. The remarkable words 

 with which he introduces this great principle give us but 

 little clue to the methods which he employed for the 

 determination of these first chemical constants, whilst in 

 no subsequent publication, as in none of the papers which 

 have come to light since his death, do we find any detailed 

 explanation of how these actual numbers were arrived at. 

 He says,f " I am nearly persuaded that the circumstance " 

 (viz., that of the different solubilities of gases in water) 

 " depends upon the weight and number of the ultimate 

 particles of the several gases : those whose particles are 

 hghtest and single being less absorbable, and the others 

 more, according as they increase in weight and complexity. 

 An inquiry into the relative weights of the ultimate particles 

 of bodies is a subject, as far as I know, entirely new. I 

 have been lately prosecuting this inquiry with remarkable 

 success. The principle cannot be entered upon in this 

 paper ; but I shall just subjoin the results, as far as they 

 appear to be ascertained by my experiments." 



Here follows the table of the relative weights of the 

 atoms. 



Table of the Relative IVeights of the Ultimate Particles 

 of Gaseous and other Bodies. 

 Hydrogen ... ... ... ... ... i 



Azot ... 4'2 



Carbon ... ... ... ... . . 4-3 



Ammonia ... ... ... ... ... 5"2 



•'''ygcn 5'5 



Water 6-5 



Phosphorus ... ... ... ... ... 7*2 



Phosphuretted liydrogen ... ... ... 8"2 



Nitrou.5 gas ... ... ... ... ... 93 



Kther 9-6 



Caseous oxide of carbon ... ... ... 9*8 



Nitrous oxide... ... ... ... ... lyj 



.Sulphur ... ... ... ... ... 14*4 



Nitric acid ... .. ... ... ... I5'2 



Sulphuretted hydrogen ... ... ... I5'4 



Carbonic acid... ... ... ... ... 15-3 



Alcohol ... ... ... ... ... i5'i 



Sulphurous acid ... ... ... ... ig-g 



Sulphuric acid ... ... ... ... 25'4 



Carburetted hydrogen from stagnant water... 63 

 Olefiant gas ... ... ... ... ... 5-3 



In the second part of his " New System of Chemical 

 Philosophy," published in iSio, Dalton points out, under 

 the description of each substance, the experimental evi- 



* By Prof H. E. Roscoe, F.RS. ; read before the Literary and Philoso- 

 phical Society of Manchester, Nov. 17, 1S74. 

 t Manch. Mem., vol. i.. Second Series, p. 286. 



