832 Dr. Adamson on Geometrical Demonstrations. 



between the cases where no nitric acid is present, and where it 

 occurs, when the substance is heated with muriatic acid and 

 the feri*ocyanide. When nitric acid is not present, the mixture 

 quickly becomes of a blue colour ; but where that acid exists, it 

 first becomes of a yellowish-green, then of an olive or dark brown ; 

 but these indications alone are not sufficient to prove the pre- 

 sence or absence of nitric acid, until afterwards confirmed by the 

 action of an alkaline sulphuret. In heating the mixture, the 

 temperature stated should be maintained for a few moments till 

 it ceases to acquire a darker shade, thereby indicating that all 

 the nitroprusside is formed. 



Sulphuric acid may be substituted for muriatic in using this 

 test; but I prefer the latter, as being more easily procured 

 pure, and as strong sulphuric acid aided by heat will decompose 

 the nitroprussides. 



This test may also admit of application to nitrous acid and 

 the nitrites ; but as those compounds are comparatively unim- 

 portant, I will not now enter on this subject. 



LIV. On a proposed Test of the Necessity of Indirect Proof in 

 Geometrical Demonstrations, with Remarks on Methods of 

 Demonstration. By James Adamson, D.D. 

 [Continued from p. 299.] 



IT does not seem difficult to discover a characteristic capable 

 of determining what are the character, relation, and effect 

 of truths of different orders employed in reasoning. All truths 

 so employed, in a definite series of arguments, are separable into 

 two divisions, according as they are, or are not, primarily made 

 use of as hypotheses. We are perfectly sure that that proposi- 

 tion is essential, which is either singly modified to become a con- 

 clusion, or which is combined with something else to constitute 

 a conclusion. Any other truths expressible in general terms, 

 can be introduced only to facilitate the transition from the one 

 form of statement, or the hypothesis, to the other form or con- 

 clusion. They therefore are auxiliary only, and the employment 

 of them individually will be contingent on the form of argument 

 which may be adopted. 



Names employed to designate such orders of truths ought to 

 classify distinctly things so different. No truth of the essential 

 or hypothesis class should rank among those of the individually 

 non-essential, or auxiliary class. The distinction does not rest 

 on a proposition being more or less certain, or more or less ob- 

 vious, but on their well-defined dissimilarity in use and value. 

 This is dependent on a difference in generahty of character ; for 



