400 



Mr. F. Galton on Blood-relationship. 



[June 13, 



problems. "We see that parents are very indirectly and only partially 

 related to their own children, and that there are two lines of connexion 

 between them, the one of large and the other of small relative importance. 

 The former is a collateral kinship and very distant, the parent being de- 

 scended through two stages (two asterisks) from a structureless source, and 

 the child (so far as that parent is concerned) through five totally distinct 

 stages from the same source ; the other, but unimportant line of connexion, 

 is direct and connects the child with the parent through two stages. We 

 shall therefore wonder that, notwithstanding the fact of an average resem- 

 blance between parent and child, the amount of individual variation should 

 not be much greater than it is, until we have realized how complete must be 

 the harmony between every variety and its environments in order that the 

 variety should be permanent. 



We also infer from the diagram how much nearer, and yet how subject to 

 variation, is the kinship between the children of the same parents ; for only 

 two stages are required to trace back their descent to a common origin, 

 which, however, proceeds from four separate streams of heredity, namely 

 the adult patent and latent elements of each of the two parents. 



An approximate notion of the nearest conceivable relationship between a 

 parent and his child may be gained by supposing an urn containing a great 

 number of balls, marked in various ways, and a handful to be drawn out of 

 them at random as a sample : this sample would represent the person of a 

 parent. Let us next suppose the sample to be examined, and a few hand- 

 fuls of new balls to be marked according to the patterns of those found in 

 the sample, and to be thrown along with them back into the urn. Now 

 let the contents of another urn, representing the influences of the other 

 parent, be mixed with those of the first. Lastly, suppose a second 

 sample to be drawn out of the combined contents of the two urns, to re- 

 present the offspring. There can be no nearer connexion justly conceived 

 to subsist between the parent and child than between the two samples ; on 

 the contrary, my diagram shows the relationship to be in reality much 

 more remote, and consisting of many consecutive stages, and therefore 

 hardly to be expressed by such simple chances. Whenever the balls in 

 the urns are much of the same pattern, the samples will be alike, but not 

 otherwise. The offspring of a mongrel stock necessarily deviate in appear- 

 ance from each other and from their parents. 



We cannot now fail to be impressed with the fallacy of reckoning inhe- 

 ritance in the usual way, from parents to offspring, using those words in 

 their popular sense of visible personalities. The span of the true here- 

 ditary link connects, as I have already insisted upon, not the parent with 

 the offspring, but the primary elements of the two, such as they existed 

 in the newly impregnated ova, whence they were respectively developed. 

 No valid excuse can be offered for not attending to this fact, on the ground 

 of our ignorance of the variety and proportionate values of the primary 



