THE CHROMOSOME NUMBERS OF THE BRITISH MENTHAE 251 



Discussion 



The above account of the chromosome numbers of the various species and hybrids 

 clearly indicates the importance of polyploidy in the genus Mentha. There are two 

 distinct polyploid series : the section Pulegium with a base number of 5, and the section 

 Mentha, with a base number of 6. The former provides a complete euploid series 

 from 10 to 40, and the latter from 24 to 132. These chromosome numbers are apparently 

 the result of both autopolyploid and allopolyploid changes in the genus. 



(a) Autopolyploidy 



Only three British species were found to have more than one chromosome 

 number, namely M. pulegium, M. longifolia and M. spicata. M. pulegium has four different 

 numbers, but no correlation between these and any morphological characters, or annual 

 and perennial habit, could be found in the available material. In M. spicata the two 

 cytotypes are probably separable by distinct though minor taxonomic characters. Further 

 attention is being devoted to them in order to determine the reliability of these characters. 

 The two cytotypes appear to be genetically isolated for no plants with an intermediate 

 number of chromosomes have been encountered even though both forms frequently 

 grow in close proximity. The occurrence of cytotypes with 36 and 48 chromosomes in 

 both M. spicata and M. /ongi/oZia suggests that ancestral types with a diploid number 

 of 24 may occur in both species. Ruttle (in Darlington & Janaki, 1945) has reported the 

 existence of such a plant in M. longifolia. As far as is known no similar plant has been 

 encountered in M. spicata but further work may well lead to its discovery. 



(b) Allopolyploidy 



In the type section, Mentha, of this genus the main cause of variation in chromosome 

 number is the frequent occurrence of hybrids. The position is further complicated by 

 amphidiploidy, the production of diploid gametes, and the occurrence, in one case, of a 

 haploid plant. The existence of amphidiploid plants has already been noted in the M. X 

 gentilis complex and in M. X smithiana. These amphidiploid mints are noticeably more 

 stable than most of the other hybrids in the genus. However stability in this case is 

 probably due to sexual sterility and reproduction by vegetative means rather than to 

 the attainment of a normal fertility and seed production. Neither of the two amphidiploid 

 forms of M. X gentilis has set seed in the plants under observation but further observa- 

 tions are required to confirm this point. M. X smithiana has been observed in the wild 

 state to produce a varying amount of seed, but this is not always well formed and no 

 information on its fertility is available. 



The production of functional diploid gametes has clearly been of importance in the 

 formation of several hybrid mints. In M. X muelleriana and the M. X gentilis and 

 M. X verticillata complexes only back-crosses produced from the unreduced hybrid 

 gametes are known. In M. X piperita both types of gamete must, on occasions, be 

 fertile because M. X citrata can arise from either type. The three remaining hybrid 

 mints which have been investigated are apparently Fj hybrids. M. X alopecuroides and 

 M. X villosonervata are probably highly sterile. Little is known about the third hybrid, 

 M. X nemorosa, but its greater variability suggests that it may be partially fertile and 

 back-cross with the parents. 



An important feature in this genus is the part played by vegetative reproduction in 

 the propagation of the species and their hybrids. By this means the effects of disorders 

 in sexual reproduction, which may cause varying degrees of sterility, have been over- 

 come. Thus hybrids, even when formed very infrequently, are maintained and dispersed, 

 and somatic variation in the sexually reproducing species is perpetuated. 



