338 PRESENT PROBLEMS IN EVOLUTION AND HEREDITY. 



hereditary organic units contained in the fertilized ovum. His con- 

 ception of heredity was derived from the study of man, and he sup- 

 ported the idea of continuity in the germ cells in order to account for 

 the law of transmission of "latent" characters; it is evident from this 

 law that only a part of the organic units of the "stirp" become "patent" 

 in the individual body; some are retained latent in the germ cells, and 

 become patent only in the next or some succeeding generation. For 

 example, the genius for natural science was "patent" in Erasmus Dar- 

 win, grandfather of the great naturalist, it was "latent" in his son, 

 and re-appeared intensified in his grandson, Charles Darwin. I have 

 eLsewhere* summed u]) as follows Galton's general results, which so 

 remarkably strengthen the "continuity" idea: We are made up, bit 

 by bit, of inherited structures, like a new building, composed of 

 fragments of an old one, one element from this progenitor, another 

 from that, although such elements are usually transmitted in groups. 

 The hereditary congenital constitution thus made up is far stronger 

 than the iufluences of environment and habit upon it. A large 

 portion of our heritage is unused, for we transmit peculiarities we 

 ourselves do not exhibit. The contributions from each ancestor can 

 be estimated in numerical proportions, which have been exactly deter- 

 mined from statistics of stature in the English race; thus the con- 

 tributions from the "patent" stature of the two parents together co'i- 

 stitute one-half while the contributions by "latent" heritage from the 

 grandparents constitute one-sixteenth, etc. One of the most important 

 demonstrations by Galton is the law of regression; this is the factor 

 of stability in race type which acts as gravitation does upon the pen- 

 dulum; if an individual or a family swing far from the average 

 characteristics of their race, and display exceptional physi(;al or 

 mental qualities, the principle or regression in heredity tends to draw 

 their offspring back to the average. 



Now how shall we distinguish regression from reversion? Very 

 clearly, 1 think; regression is the short pull which tends to draw every 

 variation and the individual as a whole back to the contemporary typ- 

 ical f(n'm, while rt'version is the long pull which draws the tyi)ical form 

 of one generjition back to the typical form of a very much earlier gen- 

 eration. These forces are evidently akin, and in the shades of trans- 

 ition from one type to another we would undoubtedly find a constant 

 diminution numerically in the recurrence of characters of the older 

 type, and thus "regressi<m " would pass insensibly into " reversion." 



Weismann has carried the idea of continuity to its extreme in his 

 simple and beautiful theory of heredity, which is founded upon the 

 postulate that there is a distinct form of protoplasm, with definite 

 chemical and molecular properties, set apart as the vehicle of inherit- 

 ance: this is the germ plasm, Tr, quite separate from the protoplasm 

 of the body cells or somatoplasm, S. Congenital characters arising in 



* Atlantic Monthhj, March, 1891, p. 359. 



