Gates, Tetraploid Mutants and Chromosome Mechanisms. 1 17 



Society 14 ). This culture showed that the plants in their adult rosette 

 stage were identical with the gigas of de Vries, though they passed 

 through an earlier rosette stage which was quite unlike gigas or 

 any other type. This year (1912) I grew at the John Innes Horti- 

 cultural Institution twelve plants from the remaining original seeds 15 ). 

 These were all, with certain exceptions to be mentioned later, 

 identical with the gigas of de Vries in all stages of their 

 development. The 1909 plants had all remained rosettes, so I 

 obtained no seeds from them, and I attribute the peculiar early 

 rosette development to the very hot, humid summer climate of 

 St. Louis. 



Subsequent inquiry from the Director of the Palermo Garden 

 elicited the information that the race had originated there several 

 years previously (apparently from a single plant) but that they had 

 subsequently lost the strain. The seeds from my culture of this 

 year are therefore probably the only ones of this strain in existence. 



The accompanying photographs (figs. 1 and 2) 16 ) show the 

 rosette and adult stages of typical individuals in my culture of 1912. 

 The pedigree numbers of certain individuals will be given for the 

 sake of convenience in future references. A preliminary examina- 

 tion of the chromosomes in typical plants shows their number to 

 be 28. I shall refer to this race as 0. gigas Italy. 



Every individual came into bloom in the English climate, the 

 seeds having been sown in January and the young rosettes planted 

 out in May. Nine of the adult plants were uniform and typical, 

 though showing certain very slight differences among themselves 

 in one or two points. Of the remainder, No. I, 9 corresponded 

 probably, to gigas oblonga, having a less stout stem, smaller leaves 

 which were nearly free from crinkling and possessed a nearly even 

 margin and a shape similar to oblonga. No. I, 12, when fully 

 developed, differed from the typical plants to a less degree, having 

 somewhat smaller leaves which were more deeply crinkled, and 

 more red on the buds (colour pattern 3 5, see Gates, 1911 b, 

 pi. 6). The rosette, however, was typical (fig. 1). The last aber- 

 rant plant, No. I, 4, differed from the type only in being distinctly 

 smaller in all its parts, though not small enough to be called a 

 dwarf. Its flowers were rather smaller than in 0. Lamarckiana. 

 This was one of the most interesting plants in the culture and will 

 be referred to again, later. Its leaves and flowers (see table II), 



14; This paper, which also deals with the remarkable range of variability observed 

 in 0. gigas, has since appeared in Trans. Linn. Soc., Botany, 8: 1 67. pis. 1 6. 



15) I am greatly indebted to the Director, Wm. Bateson, F. E. S., for fur- 

 nishing me with facilities for growing these and other plants. 



16) I am indebted to Mr. E. J. A Hard for his care in taking the photographs 

 in this paper. 



