THE LIMITS OF HEREDITY 191 



One of the interesting facts concerning " identical " 

 twins which emerges, and has already been shown in 

 other data, is (i) that they differ in their degree of 

 resemblance; (2) that they ma}^ show exceptional 

 similarity in most respects, and marked divergence 

 in a few. The former fact can probably be accounted 

 for, as explained above, on the basis of an earlier or 

 later separation of blastomeres, or of the budding 

 rudiments of separate embryos. Marked divergence 

 in certain characters was also obtained in armadillo 

 quadruplets, together with mirror- image effects. 

 The explanation is doubtless the same in both cases, 

 but is not clear at the present time, unless it be that 

 the}^ are biovular twins from parents heterozygous 

 only in these characters. A certain amount of 

 divergence must be expected even between identical 

 twins, just as the two sides of the face, or the measure- 

 ments of any bilaterally symmetrical organs of the 

 body ma}^ show differences. It is a more puzzling 

 fact that this bilateral asymmetry is itself inherited, 

 as shown b}^ identical twins whose faces show exactl}' 

 the same asymmetries (Fig. 34). 



A fact in this connection which is frequently for- 

 gotten is that the only reason why human identical 

 or monozvsfotic twins more closelv resemble each 

 other than ordinary brothers and sisters is because 

 man is usually heterozygous for so many characters 

 that the chances of two different germ-cells from the 

 same parents furnishing exactly the same combination 

 of characters is remote. Nevertheless, it is un- 

 doubtedly the fact that in some families in which both 

 parents are homozN'gous for many characters, brothers 

 or sisters ma}', and do, occasionally occur from 

 separate births, and yet resemble each other as much 

 as identical twins, except that there is always an age 

 difference. Biovular twin births in such families 

 may be expected to give rise occasionally to twins 



