xv THE PHILOSOPHY OF ZOOLOGY 643 



of the body throw off minute ultra-microscopic particles or " gem- 

 mules," and these find their way by various channels to the 

 developing reproductive cells, in which they accumulate until each 

 reproductive cell contains gemmules representing all parts of the 

 body. When development takes place each gemmule develops into 

 the part corresponding to that from which it has been derived. 



If this theory afforded a true explanation of the facts of repro- 

 duction, there would necessarily be accumulated in the ovum 

 gemmules representing not only every part of the body of the 

 adult, but also every stage in the development of the embryo, 

 and (since we see ancient ancestral characters occasionally reverted 

 to) something to represent the special peculiarities of former 

 generations. Now, it is a question if such an accumulation of 

 gemmules, each necessarily several times the size of a chemical 

 molecule, would not form a mass very much larger than an ovum. 

 Such a doctrine would, moreover, hardly appear to be necessary 

 in order to explain the facts. The accumulation in the ovum of 

 the hereditary tendencies (as we may call them) may only in part 

 take place during the lifetime of the individual : a good part of 

 them all, perhaps, except such as have been more recently 

 acquired might be contained in the ready-formed germinal 

 material handed down from previous generations. 



Against a hypothesis of pangenesis such as was formulated by 

 Darwin, the mode of reproduction of many plants tells more 

 strongly perhaps than any of the facts derived from the animal 

 kingdom. Many of the higher flowering plants, for example, are 

 capable of being propagated by means of a cutting of the stem or 

 root, or even by a leaf. As the new plant developed from the 

 cutting gives rise to flower and fruit, the cutting must contain 

 germinal matter ; and germinal matter must, therefore, be diffused 

 throughout the cells of such a plant. Pangenesis, unmodified, 

 would require that in such a case a large proportion of the ordinary 

 cells of the plant should receive gemmules derived from all parts. 



It is a moot point whether it is possible that any influence (such 

 as is presupposed in pangenesis) can pass from the cells of the 

 various parts of the body to the ova whether there can be any 

 communication of substance carrying with it tendencies to be 

 transmitted to the next generation. It is certain, however, that 

 an influence of a centrifugal character is exerted by the sexual 

 cells. The absence of ovaries or testes has, in many cases, a 

 marked effect on certain of the characters an effect on the 

 development and form of certain of the parts. This is seen not 

 only in higher animals (Mammals and Birds), but also among 

 some lower forms. In certain Crabs, for example, the presence of 

 Sacculina, a parasitic Rhizocephalan nourished at the expense of 

 the testes, which become destroyed, produces a very marked 

 alteration in some of the external features. But, while this is the 



