The doctrine of pangenesis supposes that these centers of nutrition form and throw off not only cells like themselves, but very minute granules, called gemmules, each of which is capable, under suitable circumstances, of developing into a cell like its parent. These minute granules are scattered through the system in great numbers. The essential organs of generation, the testicles in the male and the ovaries in the female, perform the task of collecting these gemmules and forming them into sets, each of which constitutes a reproductive element, and contains, in rudimentary form, a representative of every part of the individual, including the most minute peculiarities. Even more than this: It is supposed that each ovum and each zoosperm contains not only the gemmules necessary to reproduce the individuals who produced them, but also a number of gemmules which have been transmitted from the individuals' ancestors. If this theory be true,--and we can see no sound objection to it,--it is easy to understand all the problems of heredity. The gemmules must be very small indeed, but it may be suggested that the molecules of matter are smaller still, so this fact is no objection to the theory. It will be seen, then, that each spermatozoon, or zoosperm, actually contains, in an embryonic condition, every organ and tissue of the individual producing it. The same is true of the ovum. In other words, the reproductive elements are complete representatives, in miniature, of the parents, and contain all the elements for producing an offspring possessing the same peculiarities as the parents. Various modifying circumstances sufficiently explain the dissimilarities between parents and children. This theory is strikingly confirmed by the fact, previously mentioned, that in certain cases the ovum alone, a single reproductive element, may undergo a degree of development approaching very near to completion. It is supposed that fecundation is chiefly necessary to give to the gemmules the requisite amount of nourishment to insure development. As we shall see hereafter, this matter has a very important bearing upon several practical questions.