THE ORIGIN OF SPECIES 



cial production. Many methods of moderate efficiency have been intro- 

 duced. These inchide selection of bud variants with typical polyploid 

 characters; treatment of seed with temperature shocks; radiation of seed; 

 and the cutting off the stem of a plant, then selecting tetraploids from 

 among the callus shoots (shoots which develop just below the wound). 

 This method gives a yield of about 15 per cent tetraploids in the tomato. 

 In other plants, treatment of the wound with heteroauxin may be neces- 

 sary in order to get good results. But all of these methods have been made 

 obsolete by the colchicine method. Colchicine is an alkaloid drug derived 

 from the root of Colchicum autumiiale, the autumn crocus. It has been 

 known for many years that this drug interferes with the metabolism of 

 nucleic acid ( a major constituent of the chromosomes, as well as of some 

 other parts of the cell ) , but it was only in 1937 that it became known that 

 this drug is a mitotic poison. The prophase of a colchicine-influenced 

 mitosis is apparently normal, and the doubling of the chromosomes pro- 

 ceeds as usual. However, the spindle is either very defective or absent 

 entirely, and so the chromosomes, which have already been duplicated, 

 are all included in a single restitution nucleus. Thus a tetraploid condition 

 is established. Colchicine is usually used either in aqueous solution or in 

 an ointment of lanolin. In either case, a concentration of about 0.4 per cent 

 is adequate. The solution may be sprayed on flowers, or seed may be 

 soaked in it. Seedlings may be soaked, or a cut stem may be smeared with 

 the ointment, so that callus shoots will be tetraploid. Yields of 50 to 100 

 per cent tetraploid plants are not uncommon when colchicine is used. 



The Gigas Habitus. Much the most common polyploids are tetraploids, 

 with four haploid genomes ( 4n ) in the somatic chromosome complement. 

 The gametes are therefore 2n. Autotetraploids are known both in nature and 

 in experimental materials. As a matter of fact, one of the bases of the 

 mutation theory of De Vries was a mutant strain of Oenothera lamarckkina 

 ( Figure 107 ) which has turned out to be a spontaneously produced tetra- 

 ploid, as it has 28 chromosomes instead of the usual 14. This plant illus- 

 trates well a complex of characteristics which are generally, though by no 

 means universally, found in tetraploid plants. At the outset, it is consid- 

 erably larger than diploid lamarckiana. Because of this, De Vries named 

 it O. gigas, regarding it as a new species. The stems are thicker, and the 

 leaves are shorter, broader, and thicker than those of the diploid plants. 

 The most obvious physiological difterence is a slower growth rate, but 

 increased vitamin content has been reported for tetraploid tomatoes, and 

 it seems probable that the physiology of the tetraploid plant is modified as 

 much as its morphology. These traits collectively are referred to as the 

 "gigas habitus," because they so commonly characterize tetraploids. Yet 

 there are exceptions to all of them. There are tetraploids, for example, 

 which are dwarfed as compared to tlie diploid. 



It may be noted that the tetraploid Oenothera hiniarckiana could be re- 

 garded either as an autopolyploid or as an allopolyploid. Because the gigas 

 variety is derived from a single diploid parent species, it could be regarded 

 as an autotetraploid. But because the diploid lamarekiana itself is a per- 

 manent structural hybrid, the gigas variety could be regarded as an allo- 



294 



