February 23, 1922] 



NATURE 



241 



^ 



The Mechanism of Heredity. 



Bv Prof. T. H. Morgan, Columbia University, New York City, 



U.S.A. 



I. 



endel's Two Laivs of Here'^ity and their 

 Mechanism. 



AT the time when Mendel discovered his two 

 fundamental laws of heredity, no mechanism 

 was known in plants or animals that would ex- 

 plain- how such processes as those invoked by 

 him could be brought about; but between 

 1865 and 1900 (when Mendel's " Principles " 

 were recovered), the study of the ripening pro- 

 cess (maturation) of the egg and sperm-cell had 

 progressed so far that such a mechanism was 

 ready at hand. 



Mendel's first law — the law of segregation — 

 may be illustrated by the following example : A 

 tall edible pea crossed to a short pea gives tall 

 (hybrid) offspring. These, if self-fertilised, pro- 

 duce on an average three tails to one 

 short. Mendel pointed out that a very simple 

 hypothesis will account for this ratio of 3 : i in 

 the second generation (Fg). The original tall 

 parent contributes one element (T), and the 

 short parent another element (t) to the hybrid. 

 If at the time when its germ-cells mature these 

 elements separate (segregate), so that half the 

 eggs come to contain the element for tallness (T), 

 and the other half the element for shortness (t), 

 and if a similar process takes place in the pollen 

 of the hybrid (half the pollen grains bearing T 

 and half t), then chance fertilisation of any egg by 

 any pollen grain will be expected to give three 

 kinds of individuals, namely TT, Tf, tt, in the 

 ratio of 1:2:1. The first two kinds (TT and 

 Tt) will be tall plants, because the one (TT) is pure 

 for tallness, and because in the other (Tt) tallness 

 dominates shortness as seen in the hybrid. Hence 

 the second generation will be made up of three 

 tails to one short. 



The unique feature of the situation, the segrega- 

 tion in the germ-cells of the hybrid of the elements 

 derived from each parent, finds a parallel in the 

 distribution of the maternal and paternal chromo- 

 somes of the hybrid. For example : every cell of 

 the hybrid contains one chromosome (a) from one 

 parent, and one chromosome (A, the mate of the 

 former) from the other pTrent. But this con- 

 dition is not permanent in its germ-cells, for when 

 they arrive at the final ripening stage, the two 

 chromosomes (aA) come together, conjugate, and 

 then "segregate," i.e. they pass into oppo- 

 site cells. As a result, half the eggs contain 

 chromosome a, half chromosome A. They 

 behave like Mendel's pair of "characters." 

 Hence if the materials responsible for the difference 

 between T and t are carried by the members of the 

 same pair of chromosomes, A and a, they must 

 follow Mendel's first law. 



Mendel's second law applies to the independent 

 behaviour of two or more pairs of characters : the 

 NO. 2730, VOL. 109] 



members of each pair assorting independently of 

 the members of other pairs. It has been gene- 

 rally supposed by cytologists that at the ripening 

 of the germ-cells the members of the pairs of 

 chromosomes separate independently, in the same 

 way that Mendel supposed the individual pairs of 

 characters to be distributed. Proof was diffi- 

 cult to obtain from direct observation, but 

 recently this evidence has been abundantly and 

 convincingly obtained by Miss Carothers. If 

 then the chromosomes carry the materials (genes 

 or differentials) for the hereditary characters, they 

 behave in such a way as to ensure the success of 

 Mendel's second law. 



Had we only this parallelism to go upon we 

 should be justified, I think, in accepting the 

 chromosome theory of heredity as a working hypo- 

 thesis, but further evidence has been steadily 

 accumulating. It may be briefly summarised, yet 



Diploid Nuclei XX 



Gametes 



Fertilization 



Zygotes 



XY 



X 



XX 



X Y 



XY 



must be given in some detail; for it is the exact 

 correspondence between fact and theory that fur- 

 nishes the essential data for the conclusions 

 arrived at. 



(i) In some groups of animals it has been shown 

 that one pair of chromosomes (XY) acts as a 

 differential with respect to sex determination 

 (Fig. 1). The female has two like chromosomes, 

 called X and X ; the male has one X, and often 

 another chromosome called Y. Thus XX=9; 

 XY = 6. These chromosomes segregate at 

 maturation, as do the others. Every egg elimi- 

 nates one X in one of its polar bodies; half the 

 sperms are X-bearing, half Y-bearing. Any egg 

 (X) fertilised by an X-sperm = XX (9); any egg 

 (X) fertilised by a Y-sperm = XY (d). Thus sex 

 is here determined by a process that automatically 

 gives equal numbers of males and females. 



A son always gets his single X from his mother; 

 a daughter gets one X from her mother, another 

 from her father. Certain characters follow in their 

 heredity the course taken by these chromosomes. 

 For instance, if the mother is aa, and the father 

 is A, each son will be a, each daughter will be 

 aA. 



