April 15, 1920] 



NATURE 



216 



is intricate, and not one in which it is easy 

 to excite interest in a general audience. It is, 

 therefore, much to the lecturer's credit that he suc- 

 ceeded in making the subject not only intelligible, but 

 also interesting. He discussed first Froude's theory of 

 the screw, and then showed how the various factors in 

 the resulting equations had been checked by experi- 

 mental work both in the wind-channel and on the 

 "rotating arm " apparatus. Incidentally, he referred 

 to the flapping flight of birds, showed how difficult 

 it would be to imitate this, and doubted whether true 

 progress lay in this direction. Mankind had made 

 much use of the wheel in mechanism ; evolution had 

 led to the introduction of no such element in animal 

 life, in spite of its proved efficiency in its many 

 human applications. This afforded an argument 

 that man had here beaten uninstructed Nature. The 

 only flying animal which approached the aeroplane 

 in design was perhaps the beetle, which possibly used 

 its horny wing-covers as stationary planes and its 

 wings as a means of propulsion. 



The Parallaxes of Globular Clusters and 

 Spiral Nebulae. 



IT may be remembered that Dr. Charlier expressed 

 doubt as to the correctness of the enormous dis- 

 tances for globular clusters announced by Dr. Harlow 

 Shapley. Mr. Knut Lundmark, of Upsala Observa- 

 torv, undertook a re-examination of the question, 

 taking different lines of evidence from those used by 

 Dr. Shapley. His work is published in Kungl. Svenska 

 vetenskapsakadetniens Handlingar, Band 60, No. 8. 

 His data are avowedly of a much less precise character 

 than those used by Dr. Shapley, but they lead to 

 results of the same order of magnitude : — 



(i) The discussion of the proper motion of those 

 clusters for which data are available indicates a value 

 not exceeding i" per century. Accepting this maxi- 

 mum value, and combining it with the mean radial 

 vflocitv of clusters found by Prof. Slipher, Mr. Lund- 

 mark finds the distance 3000 parsecs, one-fifth of Dr. 

 Shapley 's value. 



(2) Use is made of Kapteyn's luminosity law. Van 

 Schouten has already applied this method to the 

 clusters M3, 5, 11, and 13, obtaining distances that 

 are, in the mean, twenty-eight times those of 

 Dr. Charlier and one-eighth of those of Dr. 

 Shapley. His work is here revised, estimation being 

 made of the spectral type of the stars from Dr. 

 Shapley 's observed colour-indices. The mean of 

 several independent estimations gives 6000 parsecs for 

 the distance of M3 and M13. 



(3) A rough estimate of distance is made from the 

 observed mean absolute magnitudes of stars of 

 different sfiectral types. Various assumptions are 

 made as regards the mean spectral type of the stars 

 employed. In the mean the distances found are about 

 eighty times those of Dr. Charlier, or one-third of 

 those of Dr. Shapley. 



(4) Holetschek has investigated the apparent magni- 

 tudes of several clusters regarded as single objects. 

 Mr. Lundmark shows that his values are about 

 7^ magnitudes brighter than Dr. .Shapley 's mean 

 values of the twenty-five brightest stars in the 

 respective clusters, this difference being very nearly 

 constant. 



It follows that the assumption that the absolute 

 magnitude of a cluster is constant will lead to relative 

 distances of the different clusters proportional to those 

 deduced by Dr. Shapley. 



The four lines of evidence outlined above, though 



NO. 2633, VOL. 105] 



individually weak, have cumulative force, and tend to 

 increase confidence in the accuracy of Dr. Shapley 's 

 work. 



Mr. Lundmark uses Prof. Slipher's radial velocities 

 of clusters to determine the sun's motion with regard 

 to them. He finds that its velocity is 381 km. /sec. 

 towards R..\. 320°, N. decl. 74°. He notes that both 

 the R..\. and declination of the solar apex as deter- 

 mined from stars tend to increase as fainter stars are 

 used. This is explained by a larger proportion of the 

 stars being outside the local cluster. He suggests 

 that his value is the limit to which the others are 

 tending. 



Mr. Lundmark passes on to consider the parallaxes 

 of the spiral nebulae. 



(i) Beginning with the Andromeda nebula, he 

 quotes all the directly observed measures of its 

 parallax. They are discordant, but their mean is near 

 zero. 



(2) The star density increases towards the middle 

 of the .'\ndromeda nebula, in spite of the nebulosity 

 tending to veil them. It is concluded that the nebula 

 is more distant than the non-nebular faint stars in 

 the region. A combination of the results of many 

 workers indicates a distance of 3000 parsecs for these 

 faint stars. 



(3) A combination of measured angular rotation of 

 spirals with the values of the linear rotational speed 

 given by the spectroscope has led to estimates of 

 distance somewhat greater than the last, say 4000 

 parsecs. It is further shown that the mass necessary 

 to control the rotation is lo^xsun, of the same order 

 as the estimated mass of the stellar system. 



(4) Making the rather doubtful assumption that the 

 dark curves in various nebula? have the same absolute 

 dimensions as the similar dark regions in the galaxy, 

 Wolf finds distances for various spirals r-anging from 

 10,000 to 200,000 parsecs. 



(5) Comparisons of the light curves of novae in 

 spirals with those in the galaxy, while they involve 

 several rather doubtful assumptions, give verv large 

 distances for the spirals, 200,000 parsecs being found 

 for the .Andromeda nebula. Bullialdus noted that the 

 .Andromeda nebula was exceptionally bright in the 

 year 1664. It is conjectured that a nova of 

 magnitude 5 or 6 may have appeared in it at that 

 time. 



From the above and other considerations Mr. Lund- 

 mark locates the spiral^ nebulae far beyond the galactic 

 limits, but inclines to the view that they are the star- 

 producing mechanisms of Mr. Jeans's theory rather 

 than counterparts of the galaxy. Their linear dimen- 

 sions appear to be much inferior to the latter, of 

 which our ideas have lately been enlarged by Dr. 

 Shapley 's and other researches. 



The Forestry Commission. 



WE are informed that the Forestry Commissioners 

 who were appointed on November 29 last at 

 once proceeded with the planting programme for 

 1919-20. The shortage of forest-tree seed has been 

 met to a great extent by purchases in .Austria and 

 elsewhere and by gifts from the United States and 

 Canada. .About 34,000 acres of afforestable land are 

 in course of acquisition by purchase or on lease, in 

 some cases below the market value and in others as 

 free gifts from landowners. Rather more than 10,000 

 acres are in England, of which 3/;oo are in Suffolk, 

 2760 in Devon, 1150 in Cumberland, and 1800 in 

 Northamptonshire and Bedfordshire. More than 

 5000 acres are in Ireland, of which 2000 are in Tyrone, 

 1500 in County Galway, 1500 in King's County, and 



