below or may expand into delicate cell-like structures. 

 These phenomena have been extensively studied in France, 

 but I have not seen the original papers. In the description 

 I read, it was said that gelatine should be added to the 

 ferrocyanide, but I have not found it necessary to do so. 

 In this case we are probably dealing with a semi-permeable 

 membrane of cupric ferrocyanide, so well known in con- 

 nection with osmotic pressure experiments. This membrane 

 allows the water to pass into the cavity behind the mem- 

 brane but does not allow the sugar solution to pass out, 

 and the gradual stretching of the membrane apparently 

 allows the precipitation of more ferrocyanide and so the 

 growth goes on. 



Allusion may also be made to some experiments by Prof. 

 Leduc, of Naples, said to have been exhibited at the French 

 Physical Society. The results are obtained by pouring a 

 solution of gelatine on a glass plate and adding a drop of 

 some salt solution. Very beautiful patterns are said to be 

 formed. I have tried to repeat some of the experiments 

 and with fair success. Assuming the semi-permeable 

 membrane theory as a working hypothesis, we may suppose 

 that a thin membrane of a silicate of the base is first formed 

 where the solution of salt and silicate meet. This may 

 permit the water of the silicate solution to pass in but 

 not allow i he salt solution to pass out, and the membrane 

 became stretched, as suggested above. In support of this 

 view we have the fact that coloured salts, such as ferrous 

 sulphate, cupric sulphate, cobalt nitrate, give coloured 

 growths, but do not colour the solution outside, nor give a 

 precipitate there. Manganese may beau exception. It 

 appears always to pass out sufficiently to colour the solu- 

 tion a pale brown. It is interesting to note that some 

 recent physiological work appears to indicate that man- 

 ganese salts may be exceptional in their power oi passing 

 through membranes. 



