52 



KARL B. KIUSTOFFERSON 



The difference between tliem lies mainly in the different develop- 

 ment of the leaf-lobes and the sepals; that is, in oidy two morphological 

 characters, which are both due to only one Mendelian factor with pleio- 

 tropic effect. The hybrid, further, has quite as good fertility as the 

 parents. On account of these facts 1 think it most correct to put both 

 forms in one species. The (juestion is then how this species should be 

 denominated, and which of these forms should be regarded the main 

 type and which the variety. 



From a genetical point of view it should perhaps be most correct 

 to describe M. oxyloba as the species and M. par vi flora as the variety. 



Fig. 4. Hud-sport in M. oxyloba, a l)rancli of o.ry/ota-tj'pe, b lieterozygote-type, 



c parviflora-type. 



The former has a factor with an important morj)hological effect, which 

 is absent in the latter. Further, the origin of M. pcrviflora from an indi- 

 vidual of M. oxyloba has been observed, as is mentioned above. It is 

 of course possible that M. oxyloba primarily has originated from M. 

 parviflora, for example by means of a positive mutation. 



The systematist would perhaps prefer to consider M. parviflora as 

 the si>ecies and M. oxyloba as the variety on account of the geographical 

 distribution of the two types. Both species may be endemic to the 

 Mediterranean countries. M. oxyloba is said to be confined to Cyprus and 

 Palestine. M. parviflora is now distributed in all countries round the 

 Mediterranean. It is probably to be found in most, perhaps in all 

 European countries; in Sweden it has been found occasionally. It has 



