264 



THE AMERICAN NATURALIST [Vol. LV 



species from which they have sprung. Once arisen, their 

 chances of survival would depend upon their ability to 

 complete with other forms in the struggle for existence. 

 There are no certain cases of tetraploidy known outside 

 of cultivation. It must be admitted, however, that their 

 identification would be difficult. I have showm that gigan- 

 tism is not an invariable diagnostic feature of tetraploid 

 daturas. As yet no cytological criteria of tetraploidy 

 have been established. The breeding behavior, which is 

 the only safe test, might easily be misinterpreted, as it 

 was apparently 1)y deVries in the case of the tetraploid 

 flir/as and nauella. Moreover, a suspected form must 

 show a pair of Mendelizing characters before a breeding 

 test can be applied. 



Despite the paucity of evidence in regard to the oc- 

 currence of tetraploidy in nature, the speaker believes 

 that it may have been one of the principal methods in the 

 evolution of plants. Its occurrence would furnish the bar- 

 rier between a new species and its parental form that 

 Darwin sought, and it would give a reason for the prev- 

 alence of even numbers in the counts of chromosome 

 ]);virs. I believe that a search for tetraploid forms in 

 naturo will be rewarded. Perhaps they will more likely 

 be found in horticultural i-accs propagated by vegetative 

 means. I take litis (K'c.'isioii to suggest the desirability 

 of testing lor tci rnploid) any r/i/ya.s-like plant that may 

 be found in the w'M or iindci' cultivation. We are making 

 a s])e(*ial stu(l\- ol tet ra [)loi(ly at the Station for Experi- 

 mental I^\olution and should be glad to receive plants 

 suspected of being tetraploid from any who do not care 

 to make the necessary tests themselves. 



Even if proper tests should show that few forms in 

 nature were tetraploid in the sense tliat each chromo- 

 somal set in somatic cells was composed of 4 homologous 

 niettilier-. teti-aploidy riiluht -till l)e a stage in the origin 

 of specie- with an e\ en nunilH^f of pairs of chromosomes, 

 hi the ;; t'onii- ill whieh leti-aploid i)lants have arisen 

 under obserxation, the 4 homologous chromosomes in a 



