552 



NA TURE 



[April 4, 1901 



Although not so trustworthy as the figures relating to mineral 

 output, the statistics of persons employed and of accidents in 

 mines are quite as important. The number of persons em- 

 ployed at mineral workings in 1899 throughout the world 

 amounted to 4,312,000, of which 1,635,000 were engaged in the 

 British Empire. The United Kingdom headed the list with 

 862,000 persons. Then followed Germany with 527,000, the 

 United States with 488,000, France with 302,000, Russia with 

 239,000, Austria-Hungary with 219,000, Belgium with 164,000, 

 and Japan with 133,000, Prior to the war the late South 

 African Republic employed ico,ooo miners. It appears that 

 the British Empire employs more than one-third of all the 

 persons engaged in mining and quarrying in the world. It 

 must, however, not be forgotten that published figures are far 

 from being absolutely accurate, and those cited by Dr. Le Neve 

 Foster are merely the best obtainable at the present time. As 

 an example of inaccuracy, the official returns from Ceylon give 

 1,108,306 persons employed in 1898 in mining in that island. 

 It is incredible that the mining industry of Ceylon, which is 

 comparatively insignificant as regards output, should afford 

 occupation to as many persons as are employed in mining in 

 all the other countries of the British Empire put together. 

 Such figures are utterly useless for calculating death rates, 

 and have, consequently, been discarded. The standard adopted 

 for death rates is the number of persons killed per 1000 

 employed, and a comparison of the figures in different countries 

 affords a good idea of the relative safety of the miner's occu- 

 pation. In Great Britain, in 1899, there were killed in coal 

 mines l'24, in other mines, 176, in quarries I'lg, and in 

 all mines and quarries i"26 per 1000 employed. For the 

 British Empire the average was 1*27 for coal mines and i 64 

 for metal mines, and for the world i '83 for coal mines and i "64 

 for gold mines. In foreign countries the average was 2*25 in 

 coal mines. It is evident, therefore, mining is conducted in 

 Great Britain with a far smaller risk of accident to the workers 

 than in most other countries. This gratifying result is due in 

 no small measure tc the untiring efforts made to improve the 

 conditions of mining by means of legislation and Government 

 inspection, Bennett H. Brough. 



THE MINERAL CONSTITUENTS OF DUST 

 AND SOOT FROM VARIOUS SOURCES} 



TVaORDENSKJOLD collected and described three different 

 ■*■ kinds of dust, one consisted of diatoms, a second of a 

 siliceous and apparently felspathic sand, both from the surface 

 of the ice in Greenland ; while a third consisted of sooty-looking 

 particles composed of elements invariably associated with iron 

 meteorites and of uncommon occurrence in terrestrial matter, 

 n'amely, besides metallic iron, cobalt, nickel, carbon, silicon and 

 phosphorus. He concluded that it was meteoric matter showered 

 down upon the earth, and that cosmic dust is falling imper- 

 ceptibly and continually. 



A great variety of mineral matters, including dust from various 

 sources, having been examined spectrographically by the 

 authors, they give an account of its composition. Specimens 

 which fell from the clouds were compared with those from 

 known terrestrial sources. The first comprised (i) solid matter 

 forming the nuclei of hail-stones collected during a storm on 

 April 14, 1897 ; (2) solid matter from hail and sleet collected 

 during a heavy shower from 2.30 p m. to 3 o'clock on March 

 28, 1896 ; (3) pumice from the Krakatoa eruption of 1883. 

 These were examined for Prof. J. P. O'Reilly, who had collected 

 them. (4) Dust from a dish exposed on November 16 and 17, 

 1897, in the outskirts of Dublin ; and other samples with a 

 similar origin which had fallen into porcelain dishes placed on 

 a" grass-plat in a garden. Varieties of flue-dust, (4) from Crewe 

 gas-works, (5) iron-works, (6) sulphuric acid works, and (7) 

 copper-smelting works, (8) volcanic dust from three different 

 sources, (9) soot from laundry, laboratory, kitchen and bedroom 

 chimneys. Flue-dust is characterised by the larger proportions 

 of lead, silver and copper than other varieties of dust and coal 

 ashes contain. Nickel and manganese are notably present, but 

 the most striking feature is the quantity of rubidium, gallium, 

 indium and thallium in all samples. Volcanic dust shows the 

 bands of lime and magnesia with strong spectra of the alkali 

 metals, and these are evidently its leading basic constituents. 



1 By Prof. W. N. Hartley, F.R.S., and Hugh Ramage. Abstract of a 

 ead at a meeting of the! Royal Society, February 21. 



NO. 1640, VOL. 63] 



Soot is of variable composition, not so much with respect to 

 the substances present as to the relative proportions of each in 

 any two samples. Its larger proportion of lime distinguishes it 

 from dust collected from the heavens. Nickel, manganese, 

 copper, silver and lead are constant constituents. The presence 

 of nickel is probably due to minute quantities of this element 

 being disseminated in coal, which is first converte.d by the carbon 

 monoxide produced in the fire into nickel tetracarbonyl, which 

 is naturally volatile but subsequently becomes decomposed and j 

 nickel or nickel oxide is deposited. j 



Dust from the clouds, collected either by itself or in hail, snow," 

 sleet or rain, exhibits a regularity in composition not seen in '. 

 other varieties of dust. It contains, apparently, the same pro- 

 portions of iron, nickel, calcium, copper, potassium and sodium. 

 The chief difference occurs in dust suddenly precipitated in sleet, 

 snow and hail, since lead is found in larger proportions in these, 

 and particularly so in one specimen from sleet. 



It is evident that the presence of nickel is not positive evidence 

 that the dust from the clouds comes from other than a terrestrial 

 source. 



The dust which fell on November 16 and 17, 1897, with its 

 similarity in composition to that of meteorites, its being attracted 

 by the magnet and its appearance are quite in favour of its 

 being of cosmic origin. On the other hand, in its composition 

 it is unlike volcanic dust, flue-dust or soot. 



STUDIES IN VISUAL SENSATION} 



'T'HE object of these studies is to frame if possible a scale of 

 visual sensation analogous to, and in correlation with, a 

 scale of physical luminosity. The method is the employment of 

 rotating discs. 



If a disc be divided into eleven concentric areas of equal 

 width, of which the inner is all white and the outer all black, 

 while the intervening areas have sectors giving a series of 

 10 per cent, increments of white, this gives on rotation a series 

 of grey rings between the black and white ; but they are of 

 very unequal values for sensation. While the step from black 

 to the darkest grey involves a large stride in sensation, seem- 

 ingly almost half-way towards the white, that from white to the 

 lightest grey is of no great amount, 



A contrast effect is very noticeable. Each grey annulus, 

 especially in the darker rings, is differentiated in sensation into 

 a darker moiety where it adjoins a lighter ring, and a lighter 

 moiety when it adjoins a darker ring. But although contrast 

 introduces a factor which somewhat distracts the judgment, the 

 disturbance is not sufficient to invalidate the conclusion that 

 equal, or approximately equal, increments of stimulus produce 

 increments of brightness which differ widely in value. 



By the use of slit discs on Maxwell's method the proportions 

 of white stimulus may be so adjusted as to give, say, three rings 

 intervening between white and black which do give approxim- 

 ately equal sensation steps. It is somewhat difficalt, however, 

 to estimate their value, and contrast again introduces a disturb- 

 ing element. We obtain only a first approximation to a scale 

 of sensation. Taking the black employed (admittedly only a 

 very dark grey and not an absolute black) as a zero, and calling 

 the value of the white 100 per cent,, both for sensation and 

 stimulus, we have, on the arbitrary scale thus formed, the 

 following percentages : — 



Sensation. Stimulus. 



1 



Black ring 

 Dark grey 

 Mid grey 

 Light grey 

 White ring 



II 



Here the equal increments of sensation are correlated with 

 increments of stimulus very nearly in geometrical progression. 



By interpolation a smoothed curve can be drawn through 

 the observed mid-point of 20 per cent, stimulus and translated 

 on to a disc. But this does not give a smooth increase of sensa- 

 tion from black to white through intervening greys. The value 

 of the mid-point is too high. 



Experiments with smoothed curves show that a mid-point 



1 Abstract of the Croonian Lecture delivered at the Royal Society 

 March 21 by Principal C. Lloyd Morgan, F.R.S. 



1 



