In what does this differ from the problem of modern science? It is true that Comte would limit positive science to "the study of phenomena in their orders of co-existence, resemblance, and succession," an idea which the word "positive" by no means conveys. And Tyndall asserts that "the man of science, if he confine himself within his own limits, will give no answer to the question" as to the origin of things. At the same time he admits that "he can clearly show that the present state of things _may_ be derivative."[136] The great masters of science, however, refuse to acknowledge any such arbitrary limitations. "The essence of science," says Sir William Thomson, "consists in inferring antecedent conditions, and anticipating future evolutions from phenomena which have actually come under observation."[137] If this be the _essence_ of science, then we presume that it is competent to throw some light on the primitive condition of the universe, and give some prevision of its future destiny. Did not Comte himself teach that the solar system was once all nebula, and that it will yet collapse into an exhausted and extinguished sun?[138] Is it true, then, that physical science by its inductive inference of "antecedent conditions," does really furnish a solid confirmation of the _à priori_ and native conviction of our race that the universe had a beginning? Then most assuredly even physical science is carrying us forward toward the ultimate unity of all truth--a unity which can be realized perfectly only by the constant mutual determination of _à priori_ and empirical knowledge, a synthesis and equipoise of physical and metaphysical truths.