304 



♦ KNOVV^LEDGE ♦ 



[Nov. 16, 1883. 



the stern or bow of a ship rises and sinks, the lines a A p, 

 h B q, separate more widely the farther they are carried 

 from A and B. Thus on a ship at P the distance q p 

 separating these lines might he half the height of the mast, 

 while on a similar ship at P' the corresponding distance 

 q' p' might be but a quarter of the mast's height. 



All this is so obvious that it scarcely needs dwelling 

 upon ; though it is strange how few of those who have 

 the phenomenon before their eyes even notice its nature, 

 far less consider its s.igni(icance. When observed and 

 considered a little, it constitutes one of the most effective 

 illustrations of the earth's rotundity I know of". 



But I liave in reality introduced this ocean proof for the 

 purpose of introducing another, which struck me as one of 

 the most singular in appearance of all the effects of the 

 earth's rotundity I had ever noticed. It will be observed 

 that the line a A, Fig. 20, is considerably longer than the 

 line b B, so that an object of given dimensions just on the 

 horizon for the point b would look considerably larger than 

 a similar object just on the horizon for the point a. At 



lutely level land surface) towards B and A. Then from 

 a the apparent extremities of the two lines (at the horizon) 

 would be seen at A, while from b the apparent extremi- 

 ties of the two lines would be seen at B ; and since « A is 

 greater than h B, it follows, since the lines are parallel, 

 that the apparent extremities as seen from a will be nearer 

 than as seen from b. 



Now, in travelling by rail over the prairie regions in 

 America, one occasionally passes over a perfectly direct 

 track on an unchanging level for several miles. And then 

 one can look back from the rear carriage on two parallel 

 lines just such as I have considered in the last paragraph. 

 Suppose now that standing up on the rear platform the eye is 

 as at a, Fig. 20. Then the rails will be seen as at B A, D C, 

 Fig. 2 1 , meeting the distant horizon H H' at A and C. 

 But if now, by stooping, the eye is brought to a con- 

 siderably lower position as at b, Fig. 20, then the rails will 

 be seen a.s ba, c d, Fig. 22, meeting the less distant horizon 

 hh' at a and c, — ac appearing considerably greater than AC, 

 for the simple reason that it is really the same distance 



sea one has no chance of noticing the effect of this change 

 of horizon distance, for the simple reason that there is 

 nothing to guide the eye as to the real distance of the sea- 

 line. But on land it may in certain cases be different. 



(viz. the breadth between the rails) seen much nearer. (In 

 Fig. 21, a c indicates the change of distance as compared 

 with A C). 



I have never so thoroughly /t'^i the rotundity of the earth 



Thus the prairies in America are in some places perfectly 

 level over a length and breadth of many miles. Suppose 

 now that two parallel lines were drawn from C, Fig. 20, 

 on the surface of the earth (whether sea surface or abso- 



as I did on the day (in the spring 1880) when first this illus- 

 tration occurred to me. I was on the rear platform of a train 

 running smoothly and swiftly over the prairie region of 

 Kansas. In the clear air of that region the two rails on 



