412 



NA TURE 



[August 24, 1905 



librium is attained. If the solution be kept at rest and 

 maintained at a constant temperature, tlie crystal will con- 

 tinue to concentrate the liquid around itself and to with- 

 draw solid material, until by diffusion of the impoverished 

 liquid the whole mass is ultimately reduced to saturation, 

 equilibrium is established, and the crystal ceases to grow; 

 but most saturated solutions are so viscous that a very 

 long time is required before this point is reached. Pro- 

 longed and vigorous stirring is required if the supersatur- 

 ation is to be completely relieved within, say, a day; without 

 stirring weeks may be required. 



Further, it may be possible, as is well known, to keep a 

 supersaturated solution in a sealed tube for years without 

 change; and it is also possible to start crystallisation in 

 such a liquid by dropping into it a crystal of the dis- 

 solved substance, or of one isomorphous with it, or some- 

 times by shaking it. 



But it is, perhaps, not generally known that super- 

 saturated .solutions are of two sorts. 



In 1897 Ostwald published some experiments upon super- 

 cooled liquids and supersaturated solutions, which were 

 carried out with the object of showing how extraordinarily 

 minute are the quantities of solid material capable of start- 

 ing crystallisation in such liquids, but at the same time 

 that they have a limit of size. He directed attention to the 

 radical difference which probably exists between the state 

 of a saturated solution which cannot crystallise spon- 

 taneously and that of the more strongly supersaturated 

 solution which can do so. 



The former is one in which crystallisation can either take 

 place spontaneously or can be induced bv stirring or 

 shalsing, or a variety of causes : this Ostwald calls the 

 lahile state. The latter is one in which crystallisation can 

 only take place if a solid crystal of the dissolved substance, 

 or a fragment of one, is brought into contact with the 

 liquid : this he calls the metastablc state. It is highly 

 probable that no amount of stirring or shaking, or intro- 

 duction of foreign substances, can make the metastable 

 liquid crystallise. 



Until recently no attempt to ascertain the exact limit 

 between the metastable and labile states, or even to 

 establish the existence of such a limit, had been successful, 

 and practically no attention has been paid to the difference 

 between them. Tamman, who measures the velocitv of 

 crystallisation by counting the number of the centres of 

 growth or nuclei which appear in a supersaturated solution, 

 does not recognise any distinction between the two states. 



During the present year a number of experiments carried 

 on by Miss F. Isaac and myself upon the strength of solu- 

 tions from which crystals are growing have shown that it 

 . is easy to determine the changing concentration of a cool- 

 ing solution by an optical method, to show that it passes 

 into the labile state, and to ascertain the temperature at 

 which the transition occurs. We have found, for example, 

 that a solution containing 48 per cent, of NaNO, is saturated 

 at 26°, is metastable between 26° and 16°, and crystallises 

 spontaneously below that temperature ; one containing 52 

 per cent, of NaNO, is saturated at 44°, and becomes labile 

 at 35°- 



In the metastable stale inoculation by a solid germ of the 

 dissolved substance, or of one isomorphous with it, is 

 necessary in order to cause the liquid to crystallise; in the 

 labile condition solid germs may be spontaneouslv generated 

 from the liquid. Take, for example, a test tube filled with 

 a solution of sodium-nilrate containing 48 parts of the salt 

 in 100 parts of solution, which is metastable at ordinary 

 temperatures ; if crystals make their appearance in this 

 solution it will only be because the dust of the room contains 

 minute particles of sodium-nitrate which fall into the tube, 

 or because crystals are deposited where drops have 

 evaporated near the surface, and accordingly the first 

 crystals appear at the surface of the liquid, and grow there 

 until they are large enough to fall to the bottom. I find 

 that such a solution, if enclosed in a sealed tube so as to 

 prevent access of germs and evaporation, cannot be made 

 to crystallise above the temprrature of 16°, although it is 

 supersaturated at all temperatures below 26°. 



Again, let a hoi solution of the same strength containing 



48 per cent, of the salt be allowed to cool down while being 



stirred. If dust containing NaNO, can be excluded, the 



liquid' will not crystallise until the temperature falls to 16°, 



NO. 1869, VOL. 72] 



when the solution passes from the metastable to the labile 

 condition. A cloud of nuclei will then form throughout the 

 liquid, and each will proceed to grow as a separate crystal ; 

 the immediate effect is to reduce the liquid to the meta- 

 stable state so that no more crystals are produced, but each 

 of these continues to grow from the liquid with which it 

 is in contact. 



If dust be not excluded, crystals may make their ap- 

 pearance upon the surface of the liquid and will soon sink ; 

 Ijut even though they be stirred about actively in the solu- 

 tion the liquid as a whole remains in the metastable state 

 until a temperature somewhat below 16° is reached, when 

 the labile region is entered and a cloud of new crystals 

 makes its appearance. 



It follows, therefore, that in a cooling supersaturated 

 solution, from which germs have not been excluded, there 

 are normally two periods of growth : one in which a com- 

 paratively small number of isolated crystals are growing 

 regularlv, and a subsequent period in which a shower of 

 small crystals is produced. Only if the rate of cooling be 

 sufficiently slow, or the stirring be sufficiently violent, to 

 keep the liquid in the metastable condition will there be 

 no second period, no sudden precipitation of nuclei. 



These events take place in all the aqueous solutions 

 which I have cxa:rin?d, and I am surprised that they 

 have not been discovered before. Thev afford a possible 

 explanation of two common features of igneous rocks, and 

 of slags — namely, the growth of comparatively large and 

 isolated porphyritic crystals, or phenocrysts, and the ap- 

 pearance of the same mineral at two or more different 

 periods. The origin and the arrested growth of phenocrysts 

 have generally been attributed to sudden change of tem- 

 perature, of pressure, or of hydration, and no other plausible 

 explanation has been given, although, as has been some- 

 times pointed out, they may occur in batholites where there 

 is no independent evidence of such changes. Pirsson has 

 recrgnii^ed the utter impossibility of the ordinary theory and 

 has recently suggested that each mineral has its crystallisa- 

 tion interval during which it continues to grow, and that 

 this is terminated by the increasing viscosity of the magma, 

 which checks the supply of further material to the grow- 

 ing phenocrysts and establishes new centres of crystallisa-1 

 tion. A similar explanation was adopted by Crosby for 

 the quartz-porphyry of the Blue Hills. He expresses it by 

 saying that owing to the increased viscosity the rate of 

 cooling overtaxes the molecular flow, which cannot keep 

 pace with the crystallisation. It is so difficult to find anv 

 satisfactory theory for the growth of phenocrysts that they 

 have even been attributed to the effect of earthquake 

 shocks. 



Now in a silicate magma, in all probability, the tem- 

 perature is sufficiently high to be that of the metastable 

 condition, the rate of cooling sufficiently slow to keep the 

 liquid in that condition for a considerable time, and the 

 viscosity sufficiently great to prevent the growing crystals 

 from sinking at once ; we have, therefore, all the con- 

 ditions favourable for the growth of porpliyriiic crystals ; 

 these must have generally originated throughout the liquid 

 as spontaneous nuclei if the magma entered the labile 

 state, or may have been started by inoculation or cooling 

 at the margin if the magma as a whole remained in the 

 metastable state. In the latter case suppose that further 

 somewhat sudden cooling brings the magma to the labile 

 condition, then there will be a sudden and spontaneous 

 second growth of nuclei which will not be able to attain 

 the dimensions of the porphyritic crystals; we have here 

 all the conditions necessary for a second generation of one 

 of the constituents of the rock. 



It is not necessary, therefore, to suppose that changes of 

 pressure played any vary great part in these matters. I 

 believe it will be found that considerations of temperature 

 and solubility are far more important. Similarly in the 

 case of the salt deposits van 't Hoff came to the con- 

 clusion that practically the only effect of changes of 

 pressure is to displace the temperature of formation of the 

 various compounds and not to alter their order or their 

 nature ; he estimates that this displacement is comparable 

 with that of the melting points under the same agency, 

 and in the case of the calcium-magnesium chlorides only 

 amounts to a few thousandths of a degree for one atmo- 

 sphere of pressure. 



