ISOLATING MECHANISMS AND SPECIES FORMATION 



Oenothera and Translocation Complexes. An unusual situation is 

 found in the genus Oenothera, the evening primroses. This plant came into 

 prominence in the early experiments in genetics, for this was the subject 

 of De Vries' studies during which he rediscovered Mendel's laws. Yet it 

 soon became apparent that some aspects of the genetic behavior of Oeno- 

 thera were very anomalous. A single species, O. hookeri, which is native 

 to the Pacific coast of North America, behaves like a typical plant geneti- 

 cally. It has large flowers and is normally cross-fertilized. The remainder 

 are found east of the Rocky Mountains. They are difficult to treat taxo- 

 nomically, and there is no agreement on the number of species, lumpers 

 giving as few as one at one extreme, and splitters as high as one hundred 

 at the other extreme. All are characterized by small flowers and self- 

 fertilization, and all show unusual genetic behavior. 



The distinctive genetic features of most Oenothera species are the fol- 

 lowing four. First, they show only 50 per cent fertility as compared to 

 O. hookeri. This is a result of the formation of defective seed rather than 

 of a small seed set. Second, when crossed to O. hookeri, "twin" hybrids 

 are formed; that is, there are two classes of Fi plants which differ in a 

 considerable number of traits. Third, in spite of the high degree of hetero- 

 zygosity which is demonstrated by the formation of the twin hybrids, the 

 plants breed true when selfed (their normal method of reproduction). 

 Finally, crossing over rarely enters into the results of Oenothera crosses, 

 but, when it does, it always involves large blocks of characters. The results 

 of crossing over in Oenothera are so striking that cross-over products were 

 originally interpreted as large mutations. 



All of these characteristics are understandable in terms of a series of 

 translocations in the various species of Oenothera. But first, a more thor- 

 ough discussion of the behavior of heterozygous translocations may be 

 helpful. Let us assume that two pairs of chromosomes are named, respec- 

 tively, a-a' and b'b', each letter identifying a chromosome end (Figure 

 104). At synapsis, then, normal pair formation occurs, and each pair be- 

 haves independently of the other. But now let us suppose that a translo- 

 cation occurs, so that only one normal chromosome of each pair remains, 

 and the other two now have the constitutions a*b' and b-a'. Homologous 

 point still synapses with homologous point, so that the translocated chro- 

 mosomes will bind the two tetrads together, forming a cross in the pachy- 

 tene (a particularly clear phase immediately after completion of synapsis), 

 and a ring of four chromosomes on the metaplase plate. In the division 

 of such a ring, alternate members typically go to the same pole in the 

 anaphasic movement, with the result that half of the gametes formed 

 get both of the normal chromosomes and half get both of the translocated 

 chromosomes. 



But there is no reason why only two pairs of chromosomes should be 

 involved in a translocation complex. If three chromosomes, designated as 

 a-a', b'b', and cc' are translocated so as to yield, in addition to the normal 

 chromosomes, strands of the constitution b*a', cb', and a*c', then a ring of 

 six would be formed, as illustrated in Figure 104. Again, alternate dis- 

 junction occurs at anaphase, so that half of the gametes have only un- 



283 



