Mf 



1905-] 



KNOWLEDGE & SCIENTIFIC NEWS. 



57 



each space sphere of 5 light-years radius ; or, in other 

 words, the stars are probably at an average distance of 

 5'i3 light-years apart. The survey of star distances 

 beyond this radius is, at present, so imperfect that we 

 can draw no conclusions therefrom as to the stellar 

 density in space. 



Professor Newcomb, however, says in his book on 

 " The Stars " that as far as the stars which have any 

 considerable proper motion go, they are pretty equally 

 scattered over the sky. These are the stars which are 

 presumably the nearest to us ; in fact, no measurements 

 we are likely to make will go beyond them, so that I will 

 assume as far as we have measurements the stars are at 

 intervals of 5'i3 light years apart. 



We will double the radius for each succeeding sphere, 

 and, still measuring from Greenwich, we should enclose 

 withm the boundary of each space the total number of 

 stars marked A; but, so far, I can only find parallaxes 

 for those marked B. 



Northern. Southern. 



+ 12 

 + 17 

 + 19 



We have now got as far as parallaxes will take us, 

 rather further indeed, for we can hardly place at their 

 true distances those stars added by the last sweep. Sir 

 David Gill, for example, says Canopus gives a parallax 

 of O'oo", but this only means he is sure it does not exceed 

 o-oii" (at least 296 light-years). 



It will be instructive, however, if we continue making 

 circles, and assume the stars still average 5-13 light- 

 years apart. 



At 960 light-years radius we enclose 7,007,888 stars. 

 1.9-0 ,, ,, ,, 56,623,104 ,, 



1,920 



3,840 



452,984,823 



But this is probably four times as many as the total 

 number of stars which could be photographed by pro- 

 longed exposure in the largest telescopes, and Dr. Isaac 

 Roberts has proved we thus reached the boundary of our 

 stellar universe, because all stars shown by exposures of 

 7 or 12 hours are also shown, down to the smallest mag- 

 nitudes, by only 90 minutes' exposure. The best 

 estimate I could find of the total number of stars of each 

 magnitude is by Mr. Gore(" Knowledge," 1901, p. 178). 

 He makes the total about 100,000,000 (probably about 

 70 per cent, are in the Milky Way). 



This map I have described, on which light travels one 

 mile in one year, will cover England and stretch beyond. 

 It is too big to use. I have, therefore, constructed one 

 upon a much smaller scale, fig. i ; but to realise what it 

 means the large map should always be kept in mind. 

 Even so, I have only been able to set down the stars to 

 scale as far as the 30 light-years radius; beyond that each 

 circle on my map is supposed to be twice the radius of 

 the preceding one. It will be borne in mind that, as 

 'spheres of space are represented by circles on a plane 

 surface, t^vo stars which appear near together on the 

 map may Be really far apart. The distances froin the 

 Sun, however, and from the Equator are approximately 

 correct. 



We are accustomed to think of gas and electric lights 

 as being of so many candle-power each. I have therefore 

 drawn the stars of different shapes, which distinguish 

 how many Sun-powers each star is. Within the 15 light- 

 years circle, for example, it will be found there are several 

 stars which give less light than the Sun, and some which 



give more. It would take 200 stars as bright as 

 o A 1 1677 to equal the Sun, but it would take 36 Suns 

 to give as much light as Sirius, which is 8-8 light-years 

 away. 



In the map the Sun, or a star of one Sun-power, is 

 drawn thus o, and three rays are added for a star of three 

 Sun-powers ; circles are added for tens, hundreds, &c. I 

 have however, in fig. 2, drawn the dimensions of the 

 stars compared with the Sun on the supposition that 

 equal surfaces give equal light. Probably this is true 

 only of stars of the solar type. 



Though small stars are no doubt equally abundant at 

 all distances they are not equally noticeable. On our 

 map, therefore, we shall find the Sun-power recorded 

 increases with the average distance. For example, at 

 39 light-years, we get Vega 163 Sun-powers; at 40 light- 

 years, Capella 146 Sun-powers; at 76 light years, Achernar 

 400 Sun-powers; at 136 light years, Arcturus 1989 Sun- 

 powers ; and at from 270 to 480 light years, Rigel, Spica, 

 and Deneb, from 2500 to 9200 Sun-powers, and Canopus 

 at least 21,000 Sun-powers. To make my new view of 

 the stars I searched all recent astronomical works I could 

 find for the best estimates of parallax. I gave the pre- 

 ference to those made by Gill, Elkin, and at Yale. The 

 Sun-powers I worked out upon Professor Newcomb's 

 estimate that the Sun's magnitude is 26'4 (Gore says 26"5), 

 and, as far as I could, I used the Harvard estimates of 

 stellar magnitudes. 



No doubt many of the data are uncertain, but the best 

 estimate obtainable is worth recording, though, if very 

 doubtful, I marked it (?). It is to be hoped we shall soon 

 have heliometer parallaxes for all the second magnitude 

 stars equal to those for the first, and that by photography 

 measurements will be reached for many small stars. 



A model would be more instructive than a map. What 

 could be a fitter place than the dome of St. Paul's Cathe- 

 dral ? The Sun in the centre and all the Stars hung 

 round it, as their distances become known. They would 

 be incandescent globes, and the candle power of each 

 should be proportionate to the Sun-power of the star. 



I'or a model, however, or for a small map, such as I 

 have drawn, to convey any real idea of the dimensions of 

 our stellar universe, the great world-spreading map I have 

 described should be mentally referred to; for in that the 

 Earth is actually one whole inch from the Sun and the 

 solar system no less than five feet in diameter. 



I do not mean that we should think of space coming 

 to an end at a distance of 3800 light-years ; indeed, we 

 cannot think of it coming to an end at any distance. It 

 may well be that in the infinite ocean of space there are 

 many islands of light, of which our own stellar universe 

 is one; we cannot tell, for we are far out of sight of any 

 other land. 



That our own stellar universe is limited is nearly cer- 

 tain ; for otherwise, if light-giving stars be scattered 

 throughout the whole of infinite space (unless perchance 

 they be eclipsed by the dead, dark orbs which also wander 

 there), the night would be as bright as the day, and day 

 and night the whole heavens around us would blaze with 

 an intolerable glory. 



Lantern Slides. 



We have received a supplementary list from Messrs. Newton 

 and Co. of their new lantern slides for lecture and other pur- 

 poses. These include series on Radium, Sand Figures, Rock 

 formation, Starry Heavens, Trees and Plants, and others of 

 a scientific nature. 



