All that we have actually experienced and verified is that a complex molecular aggregate is capable of being the vehicle or material basis of life; but to the question _what life is_ we have as yet no answer. Many have been the attempts to generate life _de novo_, by packing together suitable materials and keeping them pleasantly warm for a long time; but, if all germs of pre-existing life are rigorously excluded, the attempt hitherto has been a failure: so far, no life has made its appearance under observation, except from antecedent life. But, to exclude all trace of antecedent life, it is necessary not only to shut out floating germs, but to kill all germs previously existing in the material we are dealing with. This killing of previous life is usually accomplished by heat; but it has been argued that strong heat will destroy not only the life but the potentiality for life, will break up the complex aggregate on which life depends, will deprive the incubating solution not only of life but of livelihood. There is some force in the objection, and it is an illustration of the difficulty surrounding the subject. But Tyndall showed that antecedent life could be destroyed, without any very high temperature, by gentle heat periodically applied: heat insufficient to kill the germs, but sufficient to kill the hatched or developed organisms. Periodic heating enables the germs of successive ages to hatch, so to speak, and the product to be slain; and, although some each time may have reproduced germs before slaughter—eggs capable of standing the warmth—yet a succession of such warmings would ultimately be fatal to all, and that without necessarily breaking up the protoplasmic complex aggregates on the existence of which the whole vital potentiality depends. So far, however, all effort at spontaneous generation has been a failure; possibly because some essential ingredient or condition was omitted, possibly because great lapse of time was necessary. But suppose it was successful; what then? We should then be reproducing in the laboratory a process that must at some past age have occurred on the earth; for at one time the earth was certainly hot and molten and inorganic, whereas now it swarms with life. Does that show that the earth generated the life? By no means; no more than it need necessarily have generated all the gases of its atmosphere, or the meteoric dust which lies upon its snows.