122 



SCIENCE. 



[N. S. Vol. I. No. 5. 



For if every variety contributed its repre- 

 sentatives each child would on the average 

 contain, actually or potentially, twice the 

 variety and twice the number of the ele- 

 ments, whatever they may be, that were pos- 

 sessed at the same stage of its life by either 

 of its parents, four times that of any of its 

 grandparents, 1024 times as many as any 

 of its ancestors in the tenth degree and so 

 on." 



As he holds that each offspring must 

 therefore get rid, in some way, of one-half 

 the variety transmitted from its ancestors, 

 he finds an explanation of the diversity be- 

 tween individuals in the diversity of the re- 

 tained halves of then- variety. 



Each person has two parents and four 

 grandparents ; but even in a country like 

 ours, which draws its people from all quar- 

 ters of the earth, each of the eight grandpar- 

 ents is not always a distinct person ; for 

 when the parents are cousins, this number is 

 six, or five, or even four, instead of eight. 



AxQong more primitive people who stay 

 at home generation after generation, and 

 marry within the narrow circle of their 

 neighbors, a person whose ancestors have 

 transgressed none of our social laws may 

 have a minimum ancestry of only four in 

 each generation. 



The maximum ancestry and the miaimum 

 fixed by our customs are given for ten gen- 

 erations in the two lines below. 

 2-4-8-16-32-64-128-256-512-1024=2046. 

 2-4-4- 1-1 1 111 1 =38. 



Few persons who can trace their ancestiy 

 back for ten generations are descended from 

 1024 distinct persons in that generation, 

 and in all old stable communities of simple 

 folks the number is very much smaller. In 

 the long run the number of ancestors in 

 each generation is determined by the aver- 

 age sexual environment, and it is a small 

 and pretty constant number. 



All genealogy bears indirect e^ddence of 

 this familiar fact which has not been ade- 



quately recognized by students of inheri- 

 tance. 



I have made a computation from the his-" 

 tory of the people of a small island on our 

 Atlantic coast. They lead a simple life, 

 or have done so in the past, but most of the 

 men have been sailors, and have ranged 

 much farther in search of mates than agii- 

 cultural people. I have selected three per- 

 sons whose ancestrj' is recorded in detail 

 for some seven or eight generations. These 

 three persons have no parents or grandpar- 

 ents of the same name, and they would not 

 be popularly regarded as near relations, al- 

 though two of their twelve' grandparents 

 were cousins. The generations are not 

 quite parallel, and the period covered by 

 eight in one line is covered in the two others 

 by about seven, and it may be put at about 

 7^ for the three. In 7^ generations the 

 maximum ancestry for one person is 382 or 

 1146 for three persons. 



The names of 452 of them, or nearljr half, 

 are recorded, and these 452 named ances- 

 tors are not 452 distinct persons, but only 

 149; many of these in the remoter genera- 

 tions being common ancestors of all three 

 persons in many liues. If the um-ecorded 

 ancestors were interrelated in the same way 

 as they would surely be in an old commu- 

 nity, the total ancestry of the three persons 

 for 7i generations would be 378 persons 

 instead of 1146. 



Few persons know even the names of all 

 the living descendants of each of their sixty- 

 four ancestors of the sixth genei'ation, 

 and marriage with one of them is a pure 

 chance, depending on the size of the circle 

 of acquaintance and the distance to which 

 ancestors wandered. 



If a citj^ like Baltimore, where the 

 strangers to each one of us outnumber our 

 acquaintances a thousand fold, could be 

 quarantined against people from outside for 

 a thousand j^ears, each generation would be 

 much like the present one so far as known 



