Ixvi PROCEEDINGS. 



"From this wc see that molecules are not the ultimate divisions 

 of matter. The smallest sugar particles are made up of still 

 smaller ])articles ol' other things whirh do not resemble sugar, as 

 a word is composed of letters which alone do not resemble the 

 word." 



A few sentences later we have the statement: "An atom is 

 the smallest particle of an element that can enter into com- 

 bination.'' 



iS'ow, if after faithfully working in the laboratory, book in 

 hand, the high school pupil comes away with the idea that he has 

 proved the existence of atoms, and that there is much the same 

 sort of evideoce for their reality as there is for that of sugar, 

 whose fault is it? And if after a course of lalwratory training 

 of this sort, he seems to have lost forever the power of drawing 

 from simple experiments the deductions which they warrant and 

 those only, whose fault is it? 



While at the present day the atomic theory is at thi; basis of 

 the explanations of all properties of matter both in chemistry and 

 f'hysics, it has had to be confessed hitherto not only that it has 

 not been proved but that it is perhaps incapable of direct 

 experimental proof. 



To understand the difficulty one has only to realize the extreme 

 minuteness of the 'lagnitudes to be dealt with. The calculations 

 of Lord Kelvin and others have shown that, assuming the existence 

 of atoms and molecules, the atom is probably not more than one- 

 millionth of a millimeter in diameter, that is, ten million atoms 

 placed side by side so as to toucli one another would just stretch 

 across the nail of one's little finger. Similar calculations show 

 that a cubic centimeter of air, that is, a little cube each edge of 

 which measures about the width of one's little finger nail, would 

 contain under normal conditions twenty million million million 

 molecules, each of them being a group of two atoms. Professor 

 Fleming, of London, gives the following illustration : 



"We can in a good Whit worth measuring instrument detect a 

 variation in length of a metal bar equal to one-millionth of an 

 inch. This short length would i)e occupied by 25 molecules placed 

 in a row together. We can in a good microscope see a small object 



