126 



KNOWLEDGE 



[July 1, 1891. 



raeuts on the reproduction of the roclcs of a certain class 

 of meteorites. The meteorites being melted and kept for 

 some time in the liquid condition, the constituent minerals 

 befjan to crys' illise out, and finallj-, after slow coolin,<,', a 

 rock was produced having the same constituent minerals 

 as the original meteorite. Almost the only difference 

 between the meteorites and the artificial products was the 

 absence in the latter of that hreccititcil structure which 

 IVccjucntly characterises an eruptive rock which has under- 

 gone violent mechanical strains. By the employment of 

 the modern refinements of microscopic and of chemical 

 analysis, Daubree was able to establish the absolute identity 

 of the minerals contained in his artificial products with 

 those of the meteorites. The class of meteorites for which 

 the method of reproduction was found successful were 

 those containing the smallest proportion of combined 

 silica, characterised by the presence of olivine and augite 

 and by the absence of the feldspars. Except for the 

 presence of metallic iron, the mineral composition of these 

 meteorites is very similar to that of what are termed the 

 ultra-basic rocks, i.e., those the analysis of which shows 

 the smallest proportion of silica. Many basalts and other 

 lavas come under the category of ultra-basic or basic rocks. 

 Observational evidence appeared, however, to favour the 

 view that these rocks had not been produced by the purely 

 igneous method employed by Daubree, but that the action 

 of water had played an important part in their formation. 

 It was in 1878 that MM. Fouque and Levy commenced 

 the celebrated research in which they showed that the more 

 basic eruptive rocks can be reproduced in every detail of 

 mineral composition and structure by the action of heat 

 alone without invoking the aid of pressure, or the inter- 

 vention of water or any other substance not forming a 

 constituent of the rock. To appreciate the details of their 

 method it is necessary to make clear the guiding data 

 which were furnished by the study of the minute structure 

 of rocks. Some eruptive rocks are entirely composed of 

 an aggregate of perfectly crystallised minerals. One or 

 more of the constituents (in granite, the quartz) may not 

 show crystalline faces; they have presumably solidified 

 last and have been compelled to mould themselves round 

 the crystals already formed, but their structure is 

 completely crystalline, as is shown by their action on 

 polarised light. Other rocks differ from the holocrysfalline 

 in that the crystallised minerals are imbedded m a 

 vitreous or glassy matrix which scarcely affects polarised 

 light. These rocks are classed, from the character of the 

 ground mass, as glassy rocks. The most common structure 

 of eruptive rocks is that of the third class, of which the 

 ground mass has begun to crystallise before solidification, 

 but the crystallisation has only gone as far as the produc- 

 tion of iiiicniUths. These are crystals of small size, most 

 frequently microscopic, which are so far developed that the 

 determination of their species is readily effected. They 

 are seen to be grouped round the larger crystals of the rock 

 in a manner plainly indicating their later formation. It 

 appears reasonable to suppose that the microliths are 

 formed during and after the welling up of the rock, whilst 

 the formation of the large crystals may be referred to 

 a previous epoch before the disturbance of the fluid 

 mass from its subterranean position, when a condi- 

 tion of calm fusion favoured their growth and develop- 

 ment. The temperature at which they were produced 

 must be supposed to be higher than in the case of the 

 microliths. Between these two epochs of crystallisation 

 comes the eruption, during which the older crystals may 

 be rounded, worn, or broken by shock. Hall had shown 

 that to obtain a crystalline structure instead of a glassy 

 mass, it was necessary to keep the material at a temperature 



slightly above that of the melting point of the glass ; but if, 

 as appeared probable, the minerals of the different epochs of 

 crystallisation did not possess the same degree of fusibiUty, 

 it would bu necessary in order to reproduce this association 

 of minerals to maintain the materials at a series of tempera- 

 tures successively decreasing. The result of the final opera- 

 tion might be expected to be the solidification of a mass of 

 microliths of the more fusible minerals cementing together 

 the larger crystals already formed. Such was the method 

 employed by Fouque and Levy, and the result was in com- 

 plete accordance with their expectations. As an example 

 of their work, we will describe the reproduction of a basalt 

 precisely similar in character to certain basalts found in the 

 Department of Auvergne. A mixture of substances pre- 

 pared in the laboratory of the same chemical composition as 

 the rock was placed in a platinum crucible, which was 

 maintained at a white heat for forty-eight hours. A sample 

 taken out at the end of this time, and allowed to cool rapidly, 

 showed on solidifying crystals of olivine imbedded in a brown 

 coloured glassy matrix. At the end of the first forty-eight 

 hours the position of the crucible in the furnace was changed 

 so that the temperature was lowered to that corresponding 

 to a bright red heat, at which it was kept for a second period 

 of forty-eight hours. The product obtained at the end of this 

 time showed the crystals of olivine as before, but imbedded not 

 inaglass but inamatrix composed of microliths of augite and 

 of soda-lime feldspar. Among the other rocks reproduced was 

 a leucite-lava, the crystals of leucite having rounded angles 

 just as in the natural rock, showing that this pecuUarity 

 is not necessarily due, as had been supposed, to the effects 

 of disturbance after the first epoch of crystallisation. 

 The rocks produced by the methods we have described are 

 of the same character as those formed in the volcanic 

 eruptions of the present time, which belong to the class of 

 the more basic rocks. The more siliceous rocks, such as 

 granite, which contain free silica in the form of quartz, do 

 not appear to be formed under the conditions obtaining in 

 the eruptive processes which geologists have been able to 

 observe in actual operation. When the materials of the 

 acid rocks are subjected to the processes above described, 

 the minerals which crystallise out are not those of the 

 original rock, but are of different crystalline form, even 

 when they have the same chemical composition. The 

 excess of sihca remains in the imcombined state, but has 

 characters resembling those of the variety Iniown as 

 tridymite rather than those of quartz. The acid rocks 

 and their characteristic minerals (as quartz potash- 

 feldspar and soda-feldspar) have doubtless been formed 

 by processes radically different from that of simple fusion. 

 The minerals above mentioned have been reproduced by 

 the reaction of suitable materials in the presence of water, 

 at a high temperature and pressure. Hitherto it has not 

 been found possible to produce the compacted associa- 

 tion of these minerals which constitutes an acid rock. 

 Sufficient data have, however, been obtained to justify the 

 belief that at no distant date the jiroblem of the mode of 

 formation of this class of rocks will be solved by the 

 experimental method. 



One of the most important contributions to experimental 

 geology during recent years, is the discovery of Spring, 

 that pressure is capable of inducing chemical change 

 independently of its effect in raising the temperatm-e of 

 bodies. This discovery has a direct bearing on the phe- 

 nomena of mctaiiwrphism, or the bodily conversion of 

 sedimentary rocks into others of a completely different 

 character. Spring has sho'wn, that by the application of 

 great pressure, chemical combination is induced, in cases 

 where the comijound occupies a smaller volume than the 

 components, and conversely that a decomposition is brought 



