1886.] on Suspended Crystallisation. 511 



his inquiries, so far as the substances he has experimented with are 

 concerned, confirms those of Gernez, and limits the number of bodies 

 capable of acting as nuclei within very narrow limits. In fact, he 

 concludes that the only body capable of causing the crystallisation of 

 such solutions is a crystal of the substance itself, of exactly the same 

 composition as it possesses when in a state of supersaturation. Thus 

 Glauber's salt, which exists in a supersaturated solution, combined 

 with 10 parts of water, will only crystallise by the addition of a 

 crystal of the substance also containing 10 parts of water, and is 

 perfectly inactive to crystals of the same salt which contain 7 pro- 

 portions of water, or those which are anhydrous. 



As it is almost impossible to conceive that our atmosphere is 

 laden with minute particles of the many different metallic salts 

 which we are acquainted with, some hesitation in accepting such a 

 limited explanation may be excusable ; but when we consider that it 

 has been shown by Dr. Angus Smith and others that the aii*, especially 

 in the vicinity of large manufacturing towns, is filled with small 

 particles, more especially of Glauber's salt, we are not surprised that 

 solutions of this body at least crystallise at once on the removal of 

 the filtering medium. And the probability of the explanation may be 

 further strengthened by the fact that these solutions when opened 

 in the still air of country places may retain their liquid condition for 

 considerable lengths of time. 



At the same time the limit fixed by Professor Liversidge may 

 appear a somewhat narrow one, and experiments made some time ago 

 and published in the ' Journal of the Chemical Society of London ' 

 (May 1879, and September 1882), have confirmed a few experiments 

 first indicated by Gernez, and go a considerable length in showing 

 that bodies possessing not only the same crystalline form but an 

 identical chemical structure are active nuclei in causing the 

 crystallisation of supersaturated solutions. 



In these experiments two methods for the addition of the nuclei 

 intended to excite crystallisation were adopted. The first of the^e 

 consisted of a flask and bulb tube, the supersaturated solution of the 

 salt to be experimented on being placed in the flask, whilst the small 

 bulb was filled with a solution of the body intended to act as nucleus. 

 The solution in the bulb having been thoroughly boiled, the tube is 

 stoppered with cotton wool and then introduced through the centre 

 of a second cotton plug into the neck of the flask. The contents of 

 the flask are now heated, the contents of the bulb receiving a second 

 warming from the steam rising from the solution in the flask. The 

 flasks so 23repared are then allowed to stand till perfectly cold before 

 performing an experiment. When it was desired to perform an 

 experiment the solution of the body intended as nucleus contained 

 in the bulb tube is first crystallised by touching with a platinum 

 wire, or preferably by introducing a crystal of the salt contained 

 in the bulb. Crystallisation having thus taken place in the bulb, 

 and the crystal added having become enclosed in the fresh deposit. 



Vol. XI. (No. 80.) 2 l 



