Septembeb 21, 1906.] 



SCIENCE. 



373 



they possess a high degree of probability. The 

 smallest particles which they were able to 

 count appeared to have a linear dimension of 

 6 fxjx (millimicrons, or thousandths of a mi- 

 cron) . 



The extraordinary character of this achieve- 

 ment, actually counting such small particles, 

 is best realized through comparisons. The 

 smallest object visible with the best microscope 

 has a linear dimension of about i /i. or 250 /a/x. 

 The hypothetical diameter of a molecule of 

 soluble starch has been estimated by Lobry 

 de Bruyn at about 5 fifx, and of a hydrogen 

 molecule by O. E. Meyer at about 0.1 jxix. On 

 the other hand a human blood corpuscle has 

 a diameter of about 7.5 /a or 7,500 fjLfi. The 

 diagrams in the book make these comparisons 

 yet more striking. 



Some solutions contain smaller particles, as 

 is evident from the fact that the course of the 

 light can be traced, although the individual 

 particles can not be counted. Many solutions 

 appear optically empty even under this search- 

 ing test. We may wish to define- true solu- 

 tions as those which are optically empty under 

 the 'ultra apparatus,' and colloidal solutions 

 as those which are not optically empty under 

 the same conditions. But there are countless 

 intermediate steps, and these observations 

 raise the natural question whether, after all, 

 there is any difference between true and col- 

 loidal solutions, except in degree. That we 

 fail to find lack of homogeneity in what we 

 call true solutions may be due merely to the 

 limitations of our apparatus. One naturally 

 wonders if we may not some day, by these 

 methods, be able to answer, in terms of fact 

 instead of theory, those puzzling questions; 

 in what form is a substance when in solution, 

 and why are some things soluble and other 

 things practically insoluble in given solvents? 



Siedentopf and Zsigmondy consider it de- 

 sirable to coin some new words to cover the 

 new region which they have brought within 

 the range of direct observation. They sug- 

 gest that all particles smaller than | jx shall 

 be called ultra-microscopic particles or, for 

 the sake of brevity, ultra-microns. These are 

 to be subdivided into two groups, (1) those 



which may be counted in the ' ultra apparatus,' 

 to be called sub-microscopic particles, sub- 

 microns, or hypo-microns, (2) those which 

 may not be so counted, to be called amicro- 

 scopic particles or amicrons. 



The sub-microns, in water solutions, possess 

 lasting and extremely rapid oscillatory and 

 translatory motions which differ from the 

 typical Brownian motion in several particu- 

 lars. The description of this motion, illus- 

 trated with figures, is interesting. 



As is well known, we have blue solutions of 

 colloidal gold as well as red. Zsigmondy con- 

 cludes, in harmony with the majority of other 

 authorities, that there is no observable connec- 

 tion between the color and the size of the par- 

 ticles. 



Kuby glass is unquestionably to be consid- 

 ered as a supercooled colloidal solution. It is 

 possible to obtain colorless ' ruby glass ' by 

 cooling, with sufficient rapidity, the solution 

 of metallic gold in molten glass. This color- 

 less glass, upon being heated to a temperature 

 at which it softens, turns red. Zsigmondy 

 devotes a good deal of space to a theoretical 

 discussion of how the amicrons accumulate 

 at centers to form submicrons and small 

 crystals. 



A clue is obtained to the size of the pores 

 of filters by determining the size of gold par- 

 ticles which have passed through them. For 

 instance, sub-microns of gold with linear 

 dimensions of 30 /x/x passed through Maasen, 

 Pukal and Chamberland filters, which must 

 then have some pores of that size. 



The book closes with the well-justified state- 

 ment that the 'iiltra' methods form an im- 

 portant addition to, and extension of, other 

 physical and chemical methods for the study 

 of conditions which have hitherto escaped 

 direct observation. It can not be considered 

 an exhaustive review of the whole subject of 

 colloids. Lottermoser's 'Ueber anorganische 

 ColloTde' in the sixth volume of Ahren's 

 Sammlung chemischer und chemisch-tech- 

 nischer Vortrage is a more complete su m mary 

 of the literature. But Zsigmondy has given 

 us an eminently satisfactory, and most wel- 

 come treatise on the ' ultra ' methods, and the 



