♦ KNOWLEDGE ♦ 



[Au, 



, 1885. 



living iu Fear c 

 beeu called) ^ 

 surface of oui 



\\|]i(_li diiiers in-obably in tlie case of each 

 11 riuinly as life advances, and is far short 



-inns cif any one of the celestial globes, ex- 

 liu moons of Mars. It may be that if 

 r Terror (as the attendants on Mars have 

 ve might recognise the i-otundity of the 

 home, seeing that pi-obablj' neither of 

 these moons has a diameter of more than twenty miles. 

 But it is certain that no one can ajapreciate the rotundity 

 of our earth, in such sort that not merely the circum- 

 stance that the globe is rotund is recognised, but the 

 dimensions of the globe of which the region we see at 

 any moment is a part. The best pi'oof of this is found 

 in the fact that the earth's surface appears concave so 

 soon as we see any very large extent of it. As seen fi-om 

 a balloon, for instance, the earth seems like a gigantic 

 basin, the mind not being able to take in the real truth 

 that the earth is too large for the horizon to dip recog- 

 nisably even when the eye is two or three miles above 

 the earth's surface. If one could pass away from the 

 earth to distances so great that she would be visible as a 

 globe, we should still be unable to form any idea of her 

 size, — just as now the sun, moon, planets, and stars tell 

 the eye nothing of their real dimensions. 



A curious question here suggests itself : — Supposing 

 one could pass away from the earth's sui-face steadily till 

 she appeared like a globe, what would be the changes her 

 aspect would undergo ? She would certainly appear 

 concave until a great height had been attained; and as cer- 

 tainly she would eventually appear a globe as the sun and 

 moon do : but in what way, I wonder, would the ap- 

 parently concave surface pass to a manifestly convex 

 surface ? Would this happen gradually, or would the 

 conviction suddenly force itself on the mind that the 

 surface which had appeared concave was really convex? 

 There is a familiar illusion which illustrates such a 

 change as this, and seems to suggest that the change of 

 appearance would be sudden. If you look through a 

 leus, inverting the object seen, at a convex surface, it 

 apj^ears to be concave (a coin under the same conditions 

 appear.? to have all the parts which are really in relief 

 depressed) becaxise the mind recognises the evidence given 

 bj' the shadows without being conscious that this 

 evidence has been inverted by the action of the lens. 

 Now if, while the convex surface thus appears concave 

 you introrVp.ce into the field of view some object which 

 shows which way the shadows really fall — as an upright 

 pin, or the like — you find the seeming concavity at 

 once changed to convexity, the mind being unable to note 

 how the change takes place, so rapid is it. Pcssibly this 

 would be the way in which the seeming concavity of the 

 earth would change to convexity, as we passed away to 

 the distances at which the earth would appear like a 

 celestial orb. 



Illusions affecting our ideas aboat the apparent bright- 

 ness of objects are even more deceptive than those affect- 

 ing form. The French astronomer Chacornac wrote an 

 article once in explanation of the superior brightness of 

 the discs of Jupiter and Saturn near the edoe. The ex- 

 planation was ingeilin,,,. ;- ,1 ,^,,ul,^ ],,vr ] r'l,,,,., llll-MWn 



light on the nature ;): : _, -, if 



it had only chanceil I I .: :h ii, i>: !i lie 



explained had a real I \ i-i(iiri'. A-;: mii , i , In.w- 

 ever, so far are the jmrts of Jupiter ai, - , , > : the 

 edge of their discs from being bri-li ii | ,irts 



near the middle, that the precise rL\ n , i i i: > ,; , , aud 

 in quite a marked degree. I was fir.st ka lo uUi^ivi; this 

 by theoretical considerations, which seemed to suggest 

 that the light from the parts of Jupiter near the edge 



ought to be very much less than the light from the 

 middle of the planet's disc. It so chanced that just as I 

 had satisfactorily reasoned this out, I came across 

 Chacornac's article cxplaininrf why the edge is so much 

 and so obvi.mMy l,ri-i,t,.,. ti,;,„ tl,r' middle. This led me 

 to inquire wlui Ik i- t In- case nallv w ire as he supposed or 

 not. Now, to tlinsr «li,,lia\r ] >a iil a 1 1 . nt ioH to thc phcno- 



wlii.'l- -lM.^^ iliai il,r ,•,!-, ..( .\,'i \t. -'-^ .li>.' iiiust bc darker 

 than tin laai.ll.', l', .r .:•.,. i,, , !, , a ,,„i. llitr 1, h ,ks light when 





■dis( 





ci-r. 







astl„.pla,a...i 



parts ..f til.' disc It I" 



looking as dark as its 



this in realitj- proves 



middle of the disc ; _' 



suggested, therefore, 



ments on the luminos 



he should test this i 



parts of Jupiter's disc 



middle remaiucil Icn, 



planet was graJiiall_\ 



tinted darkciiiiaj ui i~ 



transparency at oia 1 



the other). The result was dcri ; , ; : ,i . 



to the evidence of the eyes. 'Ji^ i 



seemed so obviously the bri'i'lit' 

 to the absorption of the li.jlii. ii- |,,,: 

 least bright remained \i>il.lc> i..n_. -t. 

 illusion is easily exjilaiiiiil. l'>\ caiin i 

 backgroujid of the sky tic part- la ai- tlic 

 and Saturn look brighter than they reallj- 



A noteworthy illusion was passingly indicated in what 

 I have just described. I have said that a satellite some- 

 times looks as dark as its shadow close alongside. Now 

 the shadows of the satellites look black ; but tlic satellite 

 itself cannot be black. "We see then tliai tlic aiipi arance 

 of blackness does not necessarily imply re .1 l.La kia -,-. So 

 the spots on the sun look black near the middle of the 

 umbra ; yet thej^ cannot be really black there ; and indeed 

 when examined so that the effect of contrast is avoided 

 they are found to emit a considerable amount of light. 

 Another ca.vc of illusion may Lc noticed in total eclipses 

 of the Mill. Here the l.o,ly'of the ,„oon Inoks black; yet 

 ^ " velve times as bi-ightb 



than the 

 Lchter. I 



.lies, that 



the light of the 

 leans of a neutral- 

 1 almost complete 

 nijilete opacity at 

 1 o\,,etly contrary 



'' lir^t to yield 



. , : - .■. hich looked 



I M' course, the 



I with the black 



iIl'o of Jupiter 





■alie 



, for 



: the 



full upon the half of the 

 moon turned n' I her disc is 13.', times as large 



asthemooii-- _ . To my mind, one of the 



best proofs oi' tie. I i i-liMa ~- of the solar corona, is found 

 iu the seeming blackness of the moon's disc during total 

 solar eclipse. 



But the seeming whiteness of the moon's disc when 

 she is full is quite as mtich an illusion as its seeming 

 blackness when she is between th.' .in; ami us. For the 

 moon is not really white. Sle^ i- Tnueh m. re nearly 

 black. Regarding 100 as represent iul;' ]Hrfeet whiteness, 

 the average tint of the moon's surface -s^otild he repre- 

 sented by only 17. Probably the darker portions, which, 

 when she is ftiU look only slightly less white than the 

 rest, are as dark as our ]ior[)hyrics and syenites. 



Another rec:.ii,...l.N :i]i:-a ' ;: '1. . ■ ii,-' i.ri--hti;, .> i. iluit 

 which has ch e i - • i : c , , a, m tlie 



case of the llo.; >•: , i.. i ;.■ e.,e. .■. IV, ... Tl,o l.r.Mil 



above its level ; but this is a pure illusion, due to the 

 gradual diminution of the black shadows of the sur- 



