970 



SCIENCE 



[N. S. Vol. XXIX. No. 755 



At the commencement exercises of Lehigh 

 University the following announcements were 

 made: Robert W. Hall becomes lecturer on 

 forestry as well as professor of biology; Barry 

 MacNutt, assistant professor of physics, is 

 made associate professor of physics; Percy 

 Hughes, assistant professor of philosophy, 

 psychology and education, becomes professor 

 of philosophy and education in charge of the 

 department; Vahan S. Babasinian, instructor 

 in chemistry, becomes assistant professor; 

 James Hunter Wily, instructor in physics, 

 becomes assistant professor; E. J. Gilmore is 

 appointed instructor in biology. 



DISCUSSION AND CORRESPONDENCE 

 MINIMAL QUANTITIES OP FOOD PRESERVATIVES 



A CURIOUS instance of a fallacious argument 

 cast in pseudo-mathematical form appears in 

 the evidence of Dr. Harvey W. Wiley before 

 the Committee on Interstate and Foreign 

 Commerce, House of Eepresentatives, in Feb- 

 ruary, 1906. The argument is repeated in 

 more deliberate language (identical in the 

 three) in Bulletin 84, Part II., of the Bureau 

 of Chemistry, of the Department of Agricul- 

 ture (1906), at p. Y54, in Foods and their 

 Adulterations (190Y) at p. 38 and in the Pro- 

 ceedings of the American Philosophical So- 

 ciety, Vol. 47 (1908) at p. 326. As the latter 

 publications are readily accessible to the 

 scientific world, I shall quote only the informal 

 statement of the argument before the com- 

 mittee of the house : 



This is a graphic chart showing the compara- 

 tive influence of foods and preservatives ( Fig. 1 ) . 

 Of course we have to assume the data on which 

 this chart is constructed. You will understand 

 that. 



We will suppose that a normal dose of a, drug 

 is nothing. We do not need it at all. Now 

 imagine that the lethal dose of a drug — that is, 

 the dose that will kill — is 100, ana then we go 

 to work and measure at three points— at 75, at 

 50 and at 25. These are points at which we can 

 measure. We can not measure up towards the 

 right there, because the line almost coincides with 

 the basic line, and the deviation is so slight that 

 no method of measurement that we know of could 

 distinguish them. 



I omit here some reference to an error in the 

 diagram which appears to have been corrected 

 before printing. 



The lethal dose of that drug is 100. That is 

 written up there on the left. I will just trace 

 that. The normal dose of a drug in the case of a 

 person in health is zero. Then if we use a little 

 drug I can measure it here. I can measure it 



Itro 



FiQ. 1. Graphic Chart Representing the Com- 

 parative Influences of Foods and Preservatives. — 

 Wiley. 



again here (indicating) and I can measure it. 

 again here (indicating). Now from those three- 

 points I can construct a curve and calculate the 

 lethal dose, which we will assume to be 100. That 

 much drug would kill; no drug would not hurt 

 at all. 



The relative injury of a drug can be calculated 

 mathematically from a curve constructed like that 

 on experimental data, and I could tell you mathe- 

 matically, by applying the calculus there, just 

 what the hurtful value of that drug would be at 

 an infinitely small distance from zero. You have 

 doubtless, all of you, studied calculus, and you 

 know how you can integrate a vanishing function. 

 I used to know a good deal about calculus myself, 

 and I could, by integral calculus, tell you the- 

 injurious power of a drug at an infinitely small 

 distance from zero — that is, an infinitely small 

 dose. 



Now see what a contrast tnere is between a 

 food and a drug. 



The lethal dose of a food is none at all. That 

 kills you; you are starved to death. The normal 

 dose is what you eat normally, 100. I starve a 

 man, and I measure the injury which he receives. 



