9. Although, as we have said, we can give few precise directions for this cardinal process, the Selection of the Idea, in speculating on phenomena, yet there is one Rule which may have its use: it is this:--_The idea and the facts must be homogeneous_: the elementary Conceptions, into which the facts have been decomposed, must be of the same nature as the Idea by which we attempt to collect them into laws. Thus, if facts have been observed and measured by reference to space, they must be bound together by the idea of space: if we would obtain a knowledge of mechanical forces in the solar system, we must observe mechanical phenomena. Kepler erred against this rule in his attempts at obtaining physical laws of the system; for the facts which he took were the _velocities_, not the _changes of velocity_, which are really the mechanical facts. Again, there has been a transgression of this Rule committed by all chemical philosophers who have attempted to assign the relative position of the elementary particles of bodies in their component molecules. For their purpose has been to discover the _relations_ of the particles in _space_; and yet they have neglected the only facts in the constitution of bodies which have a reference to space--namely, _crystalline form_, and _optical properties_. No progress can be made in the theory of the elementary structure of bodies, {194} without making these classes of facts the main basis of our speculations. 10. The only other Rule which I have to offer on this subject, is that which I have already given:--_the Idea must be tested by the facts_. It must be tried by applying to the facts the conceptions which are derived from the idea, and not accepted till some of these succeed in giving the law of the phenomena. The justice of the suggestion cannot be known otherwise than by making the trial. If we can discover a _true law_ by employing any conceptions, the idea from which these conceptions are derived is the _right_ one; nor can there be any proof of its rightness so complete and satisfactory, as that we are by it led to a solid and permanent truth. This, however, can hardly be termed a Rule; for when we would know, to conjecture and to try the truth of our conjecture by a comparison with the facts, is the natural and obvious dictate of common sense. Supposing the Idea which we adopt, or which we would try, to be now fixed upon, we still have before us the range of many Conceptions derived from it; many Formulæ may be devised depending on the same Independent Variable, and we must now consider how our selection among these is to be made. {{195}} CHAPTER VI. GENERAL RULES FOR THE CONSTRUCTION OF THE CONCEPTION. APHORISM XXXVIII. _The Construction of the Conception very often includes, in a great measure, the Determination of the Magnitudes._ APHORISM XXXIX.