From such considerations as these have arisen two quite different views among biologists; and, while it is not our purpose to deal with disputed points, these views are so essential to our subject that they must be briefly referred to. One class of biologists adhere closely to the view already outlined, and insist for this reason that acquired variations _can not_ under any conditions be inherited. They insist that all inherited variations are congenital, and due therefore to direct variations in the germ plasm, and that all instances of seeming inheritance of acquired variations are capable of other explanation. The other school is equally insistent that there are abundant instances of the inheritance of acquired characters, claiming that these proofs are so strong as to demand their acceptance. Hence this class of biologists insist that the explanation of heredity given as a simple handing down from generation to generation of a germ plasm is not complete, and that while it is doubtless the foundation of heredity, it must be modified in some way so as to admit of the inheritance of acquired characters. There is no question that has excited such a wide interest in the biological world during the last fifteen years as this one of the inheritance of acquired characters. Until about 1884 the question was not seriously raised. Heredity was known to be a fact, and it was believed that while congenital characters are more commonly inherited, acquired characters may also frequently be handed down from generation to generation. The facts which we have noted of the continuity of germ plasm have during the last fifteen years led many biologists to deny the possibility of the latter. The debate which arose has continued vigorously, and can not be regarded as settled at the present time. One result of this debate is clear. It has been shown beyond question that while the inheritance of congenital characters is the rule, the inheritance of acquired characters is at all events unusual. At the present time many naturalists would be inclined to think that the balance of evidence indicates that under certain conditions certain kinds of acquired characters may be inherited, although this is still disputed by others. Into this discussion we cannot enter here. The reason for referring to it at all is, however, evident. We are searching for nature's method of building machines. It is perfectly clear that variations among animals and plants are the foundations of the successive steps in advance made in this machine building, but of course only such variations as can be transmitted to posterity can serve any purpose in this development. If therefore it should prove that acquired characters can not be inherited, then we should no longer be able to look upon the direct influence of the surroundings as a factor in the machine building. We should then have nothing left except the congenital variations produced by sexual