THE ORIGIN OF SPECIES 



could have contributed one or more genomes to the species being ana- 

 lyzed. On morphological grounds, Clausen and Goodspeed regarded N. 

 si/lvestris and some member of the N. tomentosa group as the most prob- 

 able parent species of N. tahacum. Examination of the chromosomes, the 

 second step in the analysis, showed that N. tahacum had 24 pairs, while 

 TSJ. sijlvestris and N. tomentosa each had 12 pairs. Thus the numbers corre- 

 sponded to the requirements of the problem. Further, the chromosomes 

 of each of the basic species resembled some of those of the polyploid 

 species in size and shape. Finally, a series of crosses was made to test the 

 actual relationship of the chromosomes. 



The genomes of N. sylvestris and N. tomentosa may be designated as 

 S and T respectively, so that the normal diploid plants will have the for- 

 mulas SS and TT. On the assumption that these genomes have been dif- 

 ferentiated by mutation and by chromosomal rearrangements since the 

 original formation of N. tahacum, this species is designated by the for- 

 mula S^S'^ T^T^. In the hybrid between sylvestris and tahacum, S' ST^, 12 

 pairs and 12 singletons are formed at meiosis. This indicates that all of the 

 chromosomes of sylvestris had a close enough homology with those of 

 tahacum to synapse normally. This hybrid is, of course, sterile because of 

 the irregular distribution of the unsynapsed T^ chromosomes to the 

 gametes. Similarly, if tomentosa is crossed to tahacum, the primary gameto- 

 cytes show 12 pairs and 12 singletons. Thus the 12 chromosomes of tomen- 

 tosa also have their homologues among the 24 chromosomes of tahacum. 

 The next step was to cross sylvestris and tomentosa. This produced a via- 

 ble but sterile plant ST, very similar to the haploid tahacum which can be 

 produced experimentally. A small amount of pairing of chromosomes did 

 occur in this hybrid— 2.5 pairs per meiotic cell was the average. This dem- 

 onstrated the fact that there is some homology between the chromosomes 

 of sylvestris and tomentosa, but not so much but what pairing frequently 

 was completely absent. This hybrid, ST, was then treated with colchicine 

 to produce the allotetraploid, SSTT. 



The resulting plant resembled N. tahacum much more closely than it 

 did either of the parent species. While it does not morphologically dupli- 

 cate any of the many known varieties of tahacum, it does not differ from 

 them more than they do from one another. And the possibility remains 

 that, if the right varieties of sylvestris and tomentosa were selected, a more 

 exact duplication of naturally occurring tahacum might be achieved. And 

 so this may be regarded as a fairly successful duplication of the natural 

 origin of one species, Nicotiana tahacum. Yet there is one disappointing 

 feature. While the artificially produced tahacum, SSTT, is fully fertile 

 when used as a pollen parent, it is female sterile. For this reason, as well 

 as others, it is suspected that another member of the tomentosa group may 

 be the actual parent of tahacum, and N. otophora appears to be a probable 

 selection. 



A more frequently quoted example of genome analysis is Miintzing's 

 study of Galeopsis tetrahit, the hemp nettle. This species has a haploid 

 number of 16, whereas most species of this genus have haploid numbers 

 of 8. This led Miintzing to suspect polyploidy. On the basis of comparison 



S02 



