424 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1961 



with them. But to recombine genes so often does not give any com- 

 bination a chance of being properly tested. Far better to let the genes 

 recombine in one sex and pass unrecombined through the other sex, 

 down the other line. It is an admirable solution : new things can be 

 made, but good things can be kept. 



In the evolution of the flies it was not therefore surprising that a 

 new type of meiosis had been developed in one sex — it had to be the 

 heterozygous sex — in which crossing-over was suppressed. The genetic 

 system would benefit from this modification. The original type of 

 meiosis, on the other hand, was the common ancestral type still found 

 in all species of plants, animals, and protista, a type of nuclear divi- 

 sion which had arisen at the origin of sexual reproduction, a type with 

 a uniform physicochemical character. 



At the time this speculative conclusion seemed to be rash. Today, 

 however, we can clearly go further. The brilliant work that is now 

 being used to reveal the genetic structure of fungi, bacteria, bacteri- 

 ophages, and other viruses makes it indeed necessary to go further. 

 We have to say now that crossing-over of gene sequences, or nucleotide 

 sequences, is the original property of all systems capable of evolution ; 

 and we may add that sexual reproduction, as we ordinarily under- 

 stand it, is the structure built around crossing-over which has made 

 the higher organisms possible [11] . 



This view turned genetics upside down. In the short term one 

 could still see fertilization as the focus of life's processes. For it is at 

 this moment in the higher plants and animals that the individual is 

 created. But in the long term the focus was shifted to the act of 

 crossing-over and the origin of the chiasma. For this is the moment 

 when, we may say, the gene is created. On this event all the processes 

 of evolution converge and from it they all diverge. 



DETERMINATION AND UNCERTAINTY 



At an early stage in the discussion of crossing-over, the opponents 

 of the chromosome theory objected that there was no visual or direct 

 evidence that chromosomes did or could cross over at meiosis. When 

 this evidence was provided they objected that there was no reason 

 why it should happen. Fortunately the mechanical reasons were by 

 this time only too evident. Chromosomes which pair as threads al- 

 ways coil around one another. Just as pairs of textile fibers spun 

 under torsion release a part of their torsion by coiling around one 

 another, so do the chromosomes. The part of their torsion released 

 is in equilibrium with the rest which is stored ; it is available to break 

 the chromatids and to untwist them to a position where their broken 

 ends can recombine in new combinations. The specificity in pairing 

 of genes and of parts of chromosomes and the observed release of tor- 



