August 9, 1894] 



NATURE 



557 



orresponding to that of the mica-triplets as arranged by Reusch. 

 iluch a spiral arrangement is shown by the paints of the corre- 

 Iponding Sohnckian system. 



i Optical Anomalies. — As already mentioned, we owe to 

 ikewster the establishment of the relation between the opiical 

 •fhavioiir of crystals and the systems of crystallisation Hut in 

 he course of his long research Brewster met with numeroin 

 liuzzling exceptions, and to the investigation of the orit^in 

 f their peculiar optical behaviour he devoted much study ; 

 ibsequent workers have concurred in expressing their admira- 

 on of the accuracy of his observations and descriptions, more 

 specially when regard is had to the extreme simplicity of the 

 pparatus available in those early days. 



It was recognised by Brewster that some of these optical 



;nomalies are due to a condition of strain, of the crystal as in 



ne case of the diamond. But in other minerals, as analcime 



nd apophyllite, the hypothesis of strain was not entertained by 



I im : he regarded the crystals as being truly composite and not 



imple ; and, recognising optically dififierent kinds of apophyllite, 



'ent so far as to give to one of them the specific name of tesse- 



te by reason of its distinctive characters. Biot, on the other 



and, sought to account for this kind of optical behaviour in 



nolher way, by the hypothesis of lamellar polarisation : a 



irystal of alum, for example; he held to be built up of thin 



Umina" arranged parallel to the octahedral planes, and imagined 



lat light which had traversed such a crystal is polarised by its 



iassage through the aggregation of laminae in the same way as 



|y passage through a pile of glass plates. But in the latter case 



here is a frequent passage of the light from air to glass and 



lass to air, whereas in the case of alum there is no evi^lence of 



lie existence of atmospheric intervals. Frankenheim sought to 



|vercome this difficulty by the further hypothesis that the 



liccessive layers of a composite posassium- and ammonium- 



lum are of different chemical composition, but such a differ- 



nce of material would be insufficient for the desired object by 



■;ason of the nearness to each oiher of the refractive indices of 



lums of different composition. Still it is a remarkable fact that 



■either a pure potassium-alum nor a pure ammonium-alum 



lows any depolarisation-effects at all ; these belong only to 



le alums of mixed composition, an<l yet there is no visible 



ifference in the physical structure of the crystals of simple and 



oraposite material. 



An epoch was made in the history of the so-called optical 



nomalies by the publication in iSySof an elaborate memoir by 



rof. Ernest Mallard of Paris, whose death last month 



leprived Mineralogy of its greatest philosopher. To make the 



losition more clear, we may take as a definite illustration the 



liineral boracile. In development of faces and magnitude of 



Ingles the crystals of this mineral are, as far as measurement 



' ith the goniometer can decide, precisely cubic in their sym- 



letry. But an apparently simple crystal of boracite, when 



)(amined in polarised light, behaves exactly like a regularly 



Dmposite body. If the crystal be a rhombic dodecahedron in 



Kternal development, all the twelve pyramids which can be 



)rmei! by drawing lines from the centre to the angular points 



re found !o be exactly similar t > each other in everything but 



rientation ; and, further, each of them has the opiical 



'laracters of a liiaxal crystal, the optic bisectrix of each indi- 



idual pyramid being perpendicular to the corresponding b,ase, 



nd thus having a different direction for each of the six pairs of 



arallel faces of the dodecahedron. Hence Mallard interred 



lat boracile belongs really, not to the cubic, but to the ortho- 



iiombic system, and that its crystallographic elemenis are so 



learly those of a cubic crystal that the molecular structure is in 



able equilibrium, not only when different molecules have their 



milar lines parallel, but also when only approximately similar 



nes have the same orientation : further, the cubic symmetry of 



le external form was regarded by him as a consequence of the 



pproximation of the crystallographic elements to those of a 



abic crystal and of the variety of orientation of the constituent 



lOlecules. Variety of orientation of constituent molecules is, 



1 fact, already recognised in the caseof ordmary interpenetrant 



■vins. The variation of optical character in different crystals 



f the same substance or different parts of the same crystal was 



len explained as being due to the variation in the number of 



lolecules belonging to each mode of orientation. 



According to another view, it was contended that a crystal 



f boracile is really cubic and simple, but that, like unannealed 



lass, it is in a state of strain related to the external form. It 



as replied that the optical characters of such unannealed glass are 



NO. 1293, VOL. 50] 



changed with the changeof strain which follows the fracture of the 

 specimen, while those of boracite are unaltered when the crystal 

 is broken. To this it was rejoined that a once compressed gum 

 retains its depolarising character unchanged on fracture of the 

 specimen, and that the same permanence may very well be a 

 character of some strained crystallised bodies. 



The controversy, however, passed to a fresh stage when it 

 was discovered that boracite becomes optically isotropic when 

 sufficiently heated, and resumes an optically composite character 

 on cooling. Mallard showed that the temperature at which the 

 change takes place is a definite one, 265° C, and that a definite 

 amount of heat is absorbed or given out during the change of 

 condition. 



It is now agreed that boracite is really dimorphous ; that 

 above 265" it is cubic in symmetry, below 265° orthorhombic : 

 the only remaining point of controversy as regards boracite 

 seems to be whether the external form owes its cubic symmetry 

 to the crystallisation having taken place at a temperature higher 

 than 265 , and therefore when tlie structure itself was truly 

 cubic — oral a temperature below 265", in which case the cubic 

 character of the form would be ascribed to the fact that the 

 orthorhombic constituent particles are so nearly cubic in their 

 dimensions that at any temperature they may by variety of 

 orientation combine to form a structure having practically cubic 

 symmetry, and naturally limiting itself by faces corresponding 

 to such a symmetry. 



In exactly the same way leucite and tridymite become re- 

 spectively optically isotropic and uniaxal when sufficiently 

 heated, and the opiical characters then correspond exactly to 

 the symmetry of the external form. 



Three years ago Dr. Brauns prepared a most useful summary 

 of the nineiy-four memoirs which had up to that time been con- 

 tributed relative to the much-discussed subject of the optical 

 anomalies of crystals, and added many new experimental results 

 which had been obtained by himself. He concludes that the 

 original view of Mallard — namely, that an optically anomalous 

 structure consists merely of differently orientated particles of 

 the same kind and of symmetry approximating to a higher 

 type — is only applicable to a very limited number of crystals, 

 such as those of prehnile ; that dimorphism is the true cause 

 in others, boracite being an example ; that in the remaining 

 minerals the cau-e is strain, which in some of them is due 10 

 foreign enclosures, as in the case of the diamond, and in others 

 is due to a molecular action between isomorphous substances, 

 as in the mixed alums and the garnets. 



Platics of Gliding. — One of the most startling of crystallo- 

 graphic discoveries was one made by Reusch, who found that 

 if a crystal of calcite is compressed in a certain way each particle 

 springs imo a new but definite position, exactly as if the crystal 

 had undergone a simple shear and the particles at the same 

 time had e.ich described a semi-somersault : a simpler method 

 of producing the same result was discovered afterwards by 

 Baumhauer. If only part of the calcite crystal is sheared, the 

 two parts of the structure itself are related to each other in the 

 same way as the two parts of a twin growth ; but in general the 

 external form is different from that of a twin, since after the 

 shearing of the material few of the faces retain their former 

 crystallographic signification. The property has since been 

 shown byiBauer, Liebisch, and more especially Mui;ge, 10 

 be a very general one ; and doubtless the so-called twin lamella; 

 met with in rock-constiluents have in many cases resulted from 

 pressure during earth-movemenls long subsequent to the epoch 

 of formation ol the crystals. .Similar lamellae have been pro- 

 duced artifically in anhydrite and some kinds of felspar by ex- 

 posure of the crystals to a high temperature. 



Ficzoclcilricity. — The most remarkable addition to our 

 knowledge of the relation of minerals and electricity has been 

 the recent discovery of the electrification produced by strain 

 (piezoelectricity). It has been shown by J. and P. Curie that 

 if a quartz-plate, with faces cut parallel to the axis and silvered 

 to make theui conductive, be strained in a certain direction, the 

 two faces either become oppositely electrified or show no signs 

 ol electrification at all, according as the faces of the plate ate 

 cut to be perpendicular to the prism-faces, or to pass through 

 the prism edges. Lord Kelvin says that this result is explicable 

 by electric eolotropy of the molecule and by nothing else, a 

 character which he had suggested for the molecule thirty-four 

 years ago : experiments conlirmalory of this hypothesis of ihe 

 permanent electrification of the molecule were made some time 

 ago by Kiecke. 



