348 Vmversity of California Publications in Agricultural Sciences [Vol. 2 



roots is questionable. The presence of several nuclei in one cell and 

 the abnormal appearance of these large cells would incline one to 

 belittle the possibility that tetraploidy here had originated by cell and 

 nuclear fusion. However, it has been pointed out that the one chromo- 

 some complex observed in a giant cell indicated that nuclear fusion 

 had taken place. We cannot, therefore, dismiss the possibility that a 

 fusion of nuclei from two cells gave rise to a cell which was tetraploid 

 and thence to tetraploid roots. It seems more likely, however, that the 

 occurrence of tetraploidy and of a root with giant multinucleate cells 

 in the same plant was merely a coincidence. 



Whatever the method of origin, the frequent occurrence of tetra- 

 ploidy in somatic tissues throws some light on two much discussed 

 questions. First, there is that of the mode of origin of diploid gametes. 

 Rosenberg (1926-27), Karpechenko (1927), and others have described 

 processes in the reduction divisions of apogamous species and inter- 

 specific hybrids by which diploid gametes are formed. The increasing 

 frequency with which tetraploidy has been recorded in root tips makes 

 it seem likely that it would be found in other tissues were they 

 examined as consistently. Its occurrence in the cells of the germinal 

 line would lead to the formation of gametes with twice the normal 

 chromosome number. This has been shown to occur in Datura where 

 tetraploid shoots were found. Presumably a smaller area might be 

 affected and only a portion of the gametes formed on one shoot might 

 be diploid. 



In the second place, the frequent occurrence of somatic tetraploidy 

 has a bearing on the origin of tetraploids and tetraploid hybrids. 

 Pnmula kewensis arose as a bud sport probably from an F^ hybrid of 

 P. verficillata and P. florihunda. It has the sum of the diploid chro- 

 mosome numbers of the parent species and Clausen and Goodspeed 



(1925) have suggested that it is a true tetraploid hybrid, the bud sport 

 having arisen by a doubling of somatic chromosomes. A similar 

 explanation with the doubling occurring immediately subsequent to 

 fertilization was suggested by these investigators to explain the origin 

 of a tetraploid hybrid between Nicotiana tahacum and N. glutinosa. 

 Rosenberg (1926) has recently proposed an explanation for the origin 

 of the tetraploid Aegilops — Triticum hybrid of Tschermak and Bleier 



(1926) which depends on the chance meeting of diploid gametes 

 formed by a "semi-heterotypic" division. In the light of the fore- 

 going facts it seems much simpler to suppose that a doubling of the 

 chromosomes took place in the fertilized egg, or in some cell of the 



