206 



NATURE 



[August 12, 1909 



of the maximum and minimum for each day, and their 

 times of occurrence, to the nearest minute ; but the actual 

 lange is not given explicitly. The space allowed to the 

 date is unnecessarily large, and it would probably be found 

 possible to add the daily range without unduly crowding 

 the figures. This would be a welcome addition. 



An interesting feature is a list of the disturbed days, 

 classified i to 4 according to the intensity. The highest 

 figure, 4, is reserved for one or two exceptional disturb- 

 ances, being applied at one or tw^o stations only to 

 October 30-31, 1903. Copies are also given, except in the 

 case of Baldwin, of the curves from a considerable number 

 of the disturbed days, the same selection being made for 

 all the stations. This is likely to prove a valuable feature. 

 Its value, however, is somewhat lessened by the fact that 

 the curves are shown on a considerably reduced scale. 

 Comparisons requiring high accuracy in the time must 

 suffer. The fact that local time is employed is also some- 

 what of an obstacle to inter-comparisons. The disturbed 

 curves are all from Eschenhagen instruments, which record 

 all the elements on one sheet. This brings before the eve 

 all that was happening simultaneously in the several 

 elements. This is a distinct advantage in the study of 

 disturbances, provided the different curves can be kept 

 distinct. Sitka, owing to its higher latitude, is exposed 

 to larger magnetic storms than the other stations, and the 

 clearness of a good many of the curves reproduced suffers 

 from crossing and confusion of the declination and hori- 

 zontal force traces. October 30-31, 1903, was naturally 

 an outstanding case of this ; but on that occasion there 

 was, besides, great loss of trace, the movements being so 

 rapid that no clear record appeared on the photographic 

 paper. The sensitiveness of the horizontal force instru- 

 ment at Sitka was reduced towards the end of 1904 to 

 about I mm. = 3 7', as compared to an average of about 

 I mm. = 1-8 7 in 1902 and 1903. The sensitiveness that 

 used to be aimed at in temperate Europe is i mm. = 5 7, 

 and we cannot but think that the reduction of sensitive- 

 ness in Sitka might with advantage go a good deal further 

 than it has yet gone. ■ Though not quite so easily effected, 

 a reduction in the sensitiveness of the declination instru- 

 ment at Sitka might also be advantageous, at least for a 

 study of the larger features of magnetic storms. The 

 device of two mirrors- adopted in the magnetographs to 

 avoid loss of trace is an alleviation if the movements are 

 slow, but if, as is frequently the case, the movements are 

 not merely large but rapid, this device may only aggravate 

 the confusion of trace. 



The volumes contain a great mass of facts, clearly 

 printed,_ presented in a readily intelligible form. Having 

 put their hand to the plough, "it is to be hoped that those 

 responsible for the work of the Coast and Geodetic Survey 

 will not turn back until simultaneous records have been 

 obtained for at least one sun-spot cycle at all the stations. 



C. Chree. 



RECENT ADVANCES IN OUR KNOWLEDGE 

 OF SILICON AND OF ITS RELATIONS 

 TO ORGANISED STRUCTURES.' 



^OT only is silicon widely diffused in nature in the 

 many forms of its oxide, but it also constitutes 

 between one-third and one-fourth of the origin.al and non- 

 sedimentary rocks— of which the solid crust of the earth 

 largely consists— in these cases being chemicallv combined 

 with oxygen and various metals, forniing natural silicates. 

 The subjoined table gives a necessarilv verv rough estimate 

 of the relative proportions in which the chief constituents 

 are present. 



Tr;E Earth's Crl-st. 



Approximate average Composition of iwii-scdimentary 



Rocks. 



Oxygen ... 



Silicon 



Aluminium 



Iron ... ... ... -7 



Lnlcmm and magnesium 6 ,, 



Alkali metals 4 ,, 



1 From a discourse delivered at the Royal Institution on Friday, May 28, 

 by Prof. J. Emerson Reynolds, F.R.S. 



NO. 2o;6, VOL. Si] 



about 47 per cent. 

 ,. 28 



The crust of the earth is, in fact, a vast assemblage of 

 silicon compounds, and the products of , their disintegra- 

 tion under the influence of water and other agents are 

 the various forms of clay, sand, and chalk which constitute 

 so large a portion of the earth's surface. 



The solid crust of the earth is actually known to us for 

 but a very few miles down — thirty at most — our deepest 

 mines being mere scratchings on its surface ; but, so far 

 as known, practically all its constituents are fully 

 oxidised, and this is probably true at much greater depths. 

 During £Eons past oxygen has been absorbed as the earth 

 cooled down, and the product is the crust on which we 

 live.' Jt is probable that the proportion of oxygen 

 diminishes awav from the surface until it disappears almost 

 wholly. What of the deeper depths? Are the compara- 

 tively light elements arranged more or less in the order 

 of density? Are we to suppose that silicon and sofne 

 carbon, aluminium, calcium, the elements chiefly com- 

 prising the crust, arc those nearer the surface, and iron, 

 copper, and tlie heavier metals nearer the centre? ' 



Until recently w'e knew little more than that the earth 

 is some Sooo miles in diameter, that its mean density is ■ 

 S-6-5-7, and that its relatively thin outer skin, ,or criist, ' 

 has .approxiipately the composition already described. 

 By a very skilfiH use of earthquake observation.s 

 Mr. R. D. Oldham has,, however, lately" 'given .jus 

 something like a glimpse within the ball, and concludes 

 from his obstrvalions lliat about five-sixths of tlie eart)t's 

 radius includes fairly Iwmogeiieous material, and that -the 

 remaining sixth at the centre consists of substances of 

 much higher density. Assuming this to be even roughly 

 true, we conclude that silicon forms probably as great -a 

 proportion of this large mass of the earth — whether 'in 

 the free state or in the for.iis of silicides — as it does of 

 the crust. , 



Having thus magnified the office of the important 

 element of which I wish to speak to you, I shall pass to 

 my next point, which is how the element can be separated 

 from quartz or other forms of the oxide, for it is never 

 met with unless .combined with oxygen in any of the 

 rocks known to us. 



I have already mentioned that quartz is. a dioxide of 

 the element — in fact it is the only known oxide — hence if 

 we remove this oxygen we should obtain free silicon. 

 This is not a very difficult matter, as it is only necessary 

 to heat a mixture of finely powdered quartz with just the 

 right proporlion of metallic magnesium. The metal com- 

 bines with the oxygen of the quartz, and forms therew-ith 

 an oxide of magnesium, while silicon remains. If the 

 material be heated in a glass vessel the moment of actual 

 reduction is marked by a bright glow, which proceeds 

 throughout the niass. When the product is thrown into 

 diluted acid the magnesium oxide is dissolved, and nearly 

 pure silicon is obtained as a soft,, dark -brown powder, 

 which is not soluble in the acid. This is not crystalline, 

 but if it be heated in an electric furnace it fuses, and on 

 cooling forms the dark crystalline substance on the table, 

 which, as you see, resembles pretty closely the graphitic 

 form of carbon, though its density is rather greater (2-6, 

 graphite being 23). 



Silicon .inalogues of Carbon Compounds. 



The points of physical resemblance between silicon and 

 carbon are of small importance compared with the much 

 deeper-rooted resemblance in chemical habits which exists 

 betw'een the two elements. This is expressed in the 

 periodic table of the elements as in the following 

 diagram : — 



Na = 23, Mg = 24, AI = 27, Si = 28, P = 3I, 8 = 32, 01 = 35-5 

 Li= 7, Be= 9, B=ii, C = i2, N = I4, = 16, F=r9 



where silicon is represented as the middle term of a period 

 of seven elements of increasing atomic weights, ju'it as 

 carbon is the middle term of the previous period. The 

 fact is, these two electro-negative or non-metallic elements 

 play leading parts in the great drama of nature, silicon 



1_ An interesting calculation has been made by Mr. Gerald Sfoney, from 

 which it appears that a stratum only g feet in depth of the surface of the 

 enrth contains as much oxveen as the whole of our present atmosphere. 

 (See P/,i7. .Vas-.. iSoo, p. M.) 



- R. D. Oldham, '"'ponstitution of the Interior of the Earth." (Quarterly 

 Journal of the Geobgical Society, vol. Ixii., 1906, pp. 456-75.) 



