Oct. 1G 1885.] 



KNOWLEDGE 



--^LV*" AN ILLUSTRATED ^ ^-J^ 



^^ MAGAZINE OF SCIENCE ^- 

 = PuinlyWorded-ExactlyDescribed 



LONDON: FEIDAY, UCTUBEB 16, 1885. 



Contents of No. i 



The Eclipsed Moou Sun 



THE ECLIPSED MOON SUNLIT. 



fur rays through each layer of the :vir above the scar 

 level. And, as an imagined observer at M could not 

 possibly get Funlight proceeding as from a focus at c or 

 c', a vague .sort of mystery spreads itself over the whole 

 inquiry. How can he get light from those parts of the 

 sun T^hich are obstinately bent on making their rays 

 focus at c or c' or anywhere except at 51, where the 

 student sujiposes that a focus is wanted (which is very 

 fir ni(l( td from being the case) ? 



Tins difficulty, or rather this bundle of difficulties, han 

 it.s orii,an m a double mistake. Firs! ilure is im; idea 

 that an atmosphere like the i:.:ii ' : ' A -..k 



if it were a spherical shell, ; 'f 



some such substance as glass; ,.: i ■ '!••■ 



idea that a spherical lens— re-\i; Jtd i- .s :i 



focus. As matters of fact :—i'ir,st, the dilicruul density 

 of the atmosphere at different heights causes it to be as 

 unlike a spherical lens, or a shell of 8nrh a \<--r.°, in i(s 

 action, as a concave lens is un'i' ■ : ' vA 



secondly, a sphere, though it !i ' )« 



passing near its centre, has ;i 'iv 



rays which pass through diffi'!;' i ■ ■ ■ ' • ►•'' 



foci are not centrally situated, like the foriis fr r.-.-jK 

 tr.iversing the sphere centrally. 



The second point does not concern us much, except in 

 so far as it serves to measure the incorrectness of the 

 ideas commonly entertained about the action of r, spbn* 

 as a lens. ^~ut the fir.st is of essential importance. 



{('o,U 



i^dfro, 



17.) 



THERE remain some points which apparently confuse 

 even " men who can render a reason " — for most of 

 their views at any rate. 



In the first place, to see an object, even in a distorted 

 form it is necessary that we should do more than merely 

 gei light from it. Rays starting from each point of an 

 object must be gathered into a focus on the retina of the 

 eye if thi'.t point is to be discerned as such ; and by the 

 combination of such focal images of the different points 

 of an object there is formed on the retina an image of the 

 object, more or less correct in its proportions. This is 

 true whether the object is looked at directly, or through 

 various media, and whether these be diffuse or fonned into 

 particular .shapes, as lenses, mirrors, and so forth. And 

 though we need not e.xpect anything like jk rfict vision in 

 the case of the sun supposed to be seen tlimujli t!ir .-irth's 

 atmosphei-e from a point on the moun's surfnci' during 

 total eclipse, yet we ought to be able to show that each 

 point of the sun would be visible as such, with suitable 

 focal adjustment of an imagined lunar eye : (the adjust- 

 ment would be different for diffei-ent points, and that 

 would prevent distinct vision of the ring-shaped image of 

 the sun as a whole, but any zone for which the eye were 

 adjusted would be visible.) 



Thu.s, suppose" E E', Fig. 4, to repr.'sent the earth, 

 S A a c, S' A' a' c the course of two rays from a point on 

 the sun throtigh the air along the curved liius A <;, A' «', 

 just grazin- the snrf:nT -f tlir .;.ril, • li,. .^i^iosite 

 points E a'Ml V. . hipI n'.-in.: :i! • :' ■ ' ■ ;i cf 



sBlc', 6-'B'///, W two IMys fmn, iIm .:,;;,,■ i~ i.,: ...-snig 



above E and E' at the same height ami crossing C c pro- 



Now, if B 6&'B' were a homogeneous sphere, the foci 

 for the rays S .\, S' A', s B, s' B', would not be on C c'. The 

 rays S .\,'s 1> ; fi > r refraction might convei-ge to a point 

 B' to a corrresponding point A', 



nf r 



at the 





:■ fn- 



Now here comes in what has proved an exasperating 

 difficulty. Some etudonts imagine tliat c (Fig. 2) is the 

 f( cus for all rays skirting the earth's atmosphere as at a 

 and a', a different focns Bomewhere along cc' resulting 



s to 



tllr 



; .\ has : 



,:d 



.ady 



S' A', than the last ray 

 of till" s.iiiiLL,' sun ytst. iday with the tii-st ray of the 

 rising sun llir day after to-morrow. 



But uuder the actual circumstances of the case, the 

 emergent rays ac,h c\ do not tend to a focus on that side 



