50 KARL B. KRISTOFFERSON 



germinated had meanwhile l)ecome too etiolated to be planted; most of 

 these plants might have been of the parviflora-lype. 



Although the segregation here recorded is a typical monohybrid 

 transgressions have been found as regards the serration of the leaves.. 

 They could not be shown with certainty in the oxyloba-lype as the mo- 

 dification is too great in this type. In some constant Fa-lines of 

 the parviflora-iype, on the other hand, they were very distinct. Some 

 7";i-lines had a much sharper serrulation of the leaves than the parvi- 

 //orfl-parent. When the serrulation was at its extreme it was very easy 

 to distinguish from the serration of the oxyîoba-lype, and even from the 

 heterozygotes. Such a transgression in a monohybrid segregation has 

 already been recorded by Nilsson-Ehle (1909) as to the black colour 

 of the glumes in oats. These transgressions depend on the presence 

 of modifying factors in either of the parents. 



Hedlund (1907) has also published a paper, which partly deals 

 with crosses between these species. He records here a genetical be- 

 haviour of the hybrids quite different to that detailed above. According 

 to a short note in »Meddelanden om Alnarps institut och egendom 1920 

 (Reports of the Institution and Estate of Alnarp) he seems to have 

 changed his opinion. There he says that these species »by crossings 

 form a monohybrid and therefore the origin of new types by means of 

 re-combination is out of question». I therefore think it quite super- 

 fluous to discuss his former interpretation. 



A rather interesting case of bud-sports was found in my cultures. 

 One plant in my pure line of M. oxyloba (fig. 4) had developed a branch 

 near the ground with leaves of the same type as the heterozygote. 

 Other branches had typical oa:y/o5a-leaves. The leaves of a branch a 

 little higher up on the main-axis were of parviflora-type. The largest 

 part of the main-axis was unfortunately destroyed by rabbits. Whether 

 this bud-sport was due to mutation or to vegetative segregation w^ill 

 probably be established by future work. The main interest at present 

 lies in the fact that the data obtained with regard to the bud-sport, as 

 well as the dominant character of the oxy/oba-characteristics indicate 

 that these latter characteristics are due to the presence of a single 

 Mendelian factor, and not to its absence. The fact that the recessive 

 (M. parviflora) has a more extended geographical distribution than the 

 dominant {M. oxyloba) points to the necessity of ascertaining the type 

 in possession of the dominant factor 0. 



M. parviflora was described as a species already by Linn.eus, M. 

 oxyloba much later, viz. 1849, which may depend on its limited distri- 



