332 



THE AMERICAN NATURALIST 



[Vol. LV 



of somatic variation we have ever observed in Nicotiana^ 

 are in. progress. The results in themselves sufficiently 

 demonstrate the existence of a genetic difference between 

 the red of macrophylla and the carmine of purpurea. 



We have also secured further data from the Cuba- 

 PuRPUEEA series which demonstrates the mode of inheri- 

 tance of carmine when crossed with the same white used 

 in the Cuba-Maceophylla series. These results are set 

 forth in Table III. The totals from the Fo populations, 



TABLE III 



F, AND Back-ckoss Data for the Cuba-Purpurea (White X Carmine) 

 Series 



c.'i inline : 4S ])iiik: 1()7 white, are to be compared with 

 a !»:;): 4 cxttccl.ition oL" 1!>4 carmine : 65 pink: 86 white. 

 The results from the back-crosses, 28 carmine: 18 pink: 53 

 white, are to be compared with an expectation based on 

 the 1:1:2 ratio of 25 carmine: 25 pink: 49 white. Pink 

 is again deficient and white in excess, but not to such an 

 extent as to give significance to the figures. Further data 

 from F.. families would be desirable for completion of 

 the analysis. Thus far the data are in agreement with 

 those presented by Allai'd for eannini' versu- pink and 



