WCP1391

[1]

16, Earlsfort Terrace,

Dublin

Nov[ember] 11 1903

Dear Sir

I am of course not in a position to criticise any other than the Astronomical parts of your new work on Man's Place in the Universe. I think parts of it are not quite correct. There is an obvious slip at p 167, though not having Newcomb's¹ book at hand I do not know the necessary correction. Instead of 295 stars with motions of 10'' a year² there is (as I think you noticed elsewhere) not a single one. I have however M. Bossert's³ Catalogue containing 2675 stars all of which have ''appreciable proper motion.'' [2] (Elsewhere I find Sir Burnham⁴ called Sir Barnham and my own initials given incorrectly but these are no doubt printer's errors).

I have never had much faith in the meteorite hypothesis. I do not believe that even in the densest swarms the meteors are packed together sufficiently closely to collide except at very rare ^long intervals. I do not think that two of those which entered our atmosphere have ever been seen to collide with each other. But supposing that the swarms are sufficiently numerous and dense to give rise to the necessary amount of collisions, space cannot I think be practically transparent as you contend even especially in the case of very distant stars. The light of these stars might probably have to pass through more than one of these dense swarms. If the [1 word illeg.] meteoritic theory be correct [1 word illeg.] the loss of light occasioned by the presence these swarms in space must I think be very considerable.

I do not think the data from Prof New [3] comb p 141 prove much. I do not believe in the reality of most of these parallaxes as I doubt whether there are 20 stars whose parallaxes are known even approximately. But that is not the point. Knowing that in the great majority of eases ^{the fainter stars} the parallaxes are too small to be measured, astronomers usually confuse themselves to trying some of those which they only suspect to be near us. Brightness is ^only one reason for entertaining this suspicion. Large proper motion is another. A binary pair with a comparatively short period is a third, and so on. There ^{has been} to no attempt to measure the parallaxes of all stars of any magnitude below the first or rather the second magnitude or even to try it for ^{a number of} average stars of lower magnitudes. With these lower stars ^magnitudes astronomers pick out and try some [1 word illeg.] because they suspect them of having considerable parallaxes. Moreover in Ball's⁵ researches (and probably others are of the same ^kind [1 word illeg.], but he presented me with his book) when after some preliminary observations the suspected star that he was trying gave no of having an appreciable parallax, he left off without arriving at any [4] precise figure. It was not worth the trouble of computing when the resulting parallax was not likely to exceed the limits of error. Pritchard⁶ tried to compute the parallaxes of all the <second?> magnitude stars visible in this country and I think his result averaged considerably less than that arrived at for those of the first magnitude. The destruction which you notice between Sirian and solar stars will ^{help to} explain why some stars of a given magnitude have much greater parallaxes than others. But most ^computed parallaxes are relative parallaxes. They mean the difference in parallax between the star in question ^and some small stars in the neighborhood whose parallax is assumed to be practically nil. In this case a parallax of 0 (if reliable) would mean that the little stars are at the same distance as the big one, and a [5] [Number 2 written here as part of original MS] negative parallax would mean that the little stars were nearer to us than the big one. But the reason why the faint stars whose parallaxes have been measured are nearer to us (on the average) than the bright ones is simply I apprehend that we only measure faint stars that are suspected of having large parallaxes and if at an early stage we have reason to conclu^de that the suspicion was unfounded the computation is never completed, as it is in the case of a bright star. But the ^{the nearest star} nearest star to us is the bright a Centauri and I think ^both Sirius and procyon ^procyon come into [6] the first ten.

In the passage cited p. 144 I believe I made an error, the correction of which would strengthen your argument. The light of the full moon was I believe compared not with the light of two the two solar hemispheres but with that of the near hemisphere of the sun only — so that all my figures should have been doubted. Moreover some rate the superiority of sun-light as high as 800,000 to 1.

You do not seem to notice that Newcomb's [1 word illeg.] ratio of 3.5 to 1 for stars of successive magnitudes, though higher than I deduced from a comparison with the Harvard Photometry, affords an argument in favor of your view (assuming space to be almost ^completely trans [7] parent). The scale of magnitudes now adopted is 2.512 to 1 for the light-ratio of each successive magnitude (this figure being adopted for convenience of logarithmic computation 10[MS illeg]⁴). Now on the hypothesis of uniform distribution this would give a ratio of 3.984 or in round numbers 4 to 1 instead of 3 1/2. The stars are therefore thinning from the outset which looks like a central position. But I doubt if the real rates, with photometrically measured magnitudes, exceeds 3 to 1 and I think Newcomb's⁷ figures p 152 are too large.

Some doubtful figures [1 word illeg.] are given in your book which are not essential to the argument, e.g[.] the period of 1000 ^years for Castor, the brightness of [1 word illeg.] Leonis as 300 times that of the sun [8] (both derived from computed orbits that do not agree well with recent measurements) and the statement the Canopus is a star of Sirian type. You do not seem aware of Sir R Ball's theory of meteorites⁸ (i.e. stones which fall in the solid state to the earth instead of being dissipated high up in the atmosphere). He thinks they were projected from terrestrial volcanoes with sufficient force⁹ to prevent them from falling back at a time when these volcanoes were much more powerful^{10 than} at present. This would afford a very simple explanation of why meteorites contain no ingredients that are not found in the earth.

The book will probably soon pass to [1 word illeg.] a second edition when some of these remarks may possibly prove useful[.]

Very sincerely yours | W H L Monck [signature]