232 



THE BEHAVIOUR OF POLYPEOIDS 



embracing records of 2,413 species. The most abundant con- 

 tributors have not always been the most exact, but there are 

 probably less than 5 per cent, of erroneous observations included. 



In order to find out the relative importance of polyploidy and 

 other changes in chromosome number, these records may be 





^^ 



Fig. 79. — Mitotic metaphases in Aconituni {x = 8) diploid and 

 tetraploid species, a triploid clone and an octoploid giant variety. 

 The chromosome types are reduplicated in the forms with 

 higher numbers. 2x [i] A. barbata. 2X (ii) A. vulparia. jx 

 A. Napellus, " Spark's Variety." 4X A. anglicum. 8x A. 

 Wilsoni. X 2000 (D., unpublished ; cf. Schafer and La 

 Cour, 1934)- 



examined from a different point of view. There are certain numbers 

 between 6 and 12 and the number 13 which can but rarely be regarded 

 as having arisen from recent polyploidy. The number of species 

 of each of these classes proves to have a definite relationship with 

 the number of those having double the number. Fig. 77 represents 

 graphically the results of the comparison. Three inferences can be 



