370 



GROWTH OF PLANTS 



show the results in the form of a three-dimensional solid model, as illustrated 

 in Fig. 144. The concept of a "toxicity surface," as is seen in the figure, 

 was thus introduced.-^ It was further sho\vn that the form of the toxicity 

 surface determines the precision of a toxicity experiment. Whether com- 

 parisons of toxic agents are made on the basis of the times required for an 

 equal percentage response, or on the basis of percentage responding in equal 

 times is largely a matter of convenience, for at a given point on the toxicity 

 surface both methods are capable of equal precision. 



Figure 144. The toxicity surface for the action of copper sulfate on conidia of 

 Sclerotinia fr u cticola. 



The dosage-response curve. The germination of fimgus spores in the 

 presence of toxic chemicals was sho\\Ti to be similar to that of toxicity 

 curves in other fields and in general to give a normal distribution when 

 plotted against the logarithm of the dose or concentration.^^ In making 

 these calculations the newly devised "normal equivalent deviations" 

 of Gaddum ^ were at first employed ^^ and then succeeded by the later 

 "probits" of Bliss ^ which are fundamentally similar and now more gen- 

 erally used. However, it was further sho^vn ^^' ^^ that in most cases rapid 

 graphic methods of calculation will suffice and for this purpose the now 

 widely used logarithmic probability paper (as may be seen in Fig. 145) 

 was introduced into the fungicide field. Here spore germination or toxicity 

 curves usually plot as straight lines and comparisons may be readily made, 

 ordinarily at the most precise midpoint of 50 per cent response, that is, 

 the LD50. 



Extensive studies have been made on the factors causing variation in 



