HEREDITY 493 



also render it practically certain that the part concerned with 

 heredity is the nucleus, i.e. the specialized particle of protoplasm 

 which every germ-cell contains. Weismann limits the field still 

 further, and considers that the nucleus is in part composed of 

 "germ plasma", a protoplasmic material specially concerned with 

 the transmission of characters. In typical egg-propagation (see 

 vol. iii, p. 335) germ-cells from the two parents fuse together, 

 and the essential point about the process seems to be the union 

 of the two nuclei. This has undoubtedly an important bearing 

 on the question of heredity, but precisely what bearing is still a 

 matter of doubt. It is perhaps the most remarkable fact in the 

 whole realm of knowledge that the fusion of two microscopic 

 particles of protoplasm should carry with it so vast a range of 

 possibilities as regards inheritance. 



There are some clear cases which prove that the germ-cells 

 are influenced by some of the factors in the surroundings. Yung, 

 for example, by bringing up tadpoles on specially nutritious food, 

 was able to produce with certainty an abnormal proportion of 

 females (90 per cent or even more), and we have elsewhere seen 

 (see p. 256) that a fertilized bee's egg may give rise to either a 

 worker or queen, according to the nature of the food received 

 by the larva. Even more remarkable is the case of certain lowly 

 crustaceans upon which Schmankewitsch experimented. In the 

 course of several generations he was able to convert a species 

 (Artemia Milhausenii) living in saltish water into another species 

 (A. salina), by gradually increasing the amount of salt. He also 

 found it possible to conduct the experiment in the reverse order, 

 and in this instance was able to go a step further, obtaining a 

 third species of a distinct genus (Branchipus stagnalis), char- 

 acteristic of perfectly fresh water. In the light of such facts it 

 seems difficult to believe that there is no possibility of acquired 

 somatic characters being transmissible, for we can scarcely main- 

 tain that in all the cases cited the germ -cells were directly 

 influenced by modification in the surroundings. 



Galton has formulated a law (since modified by Karl Pearson) 

 expressing numerically the influence of parents and remoter an- 

 cestors upon the characters of offspring, and the application of 

 mathematical methods to biological statistics is likely to yield 

 important results in the immediate future, as regards heredity, 

 variation, and many other problems. On the botanical side very 



