620 ON PHYSIOLOGICAL CHROMOMERES. 



although externally wholly different, were 0. mut. cana, pallescens, 

 liquida and spathulata. It has been assumed by M. Boedijn and 

 myself, that each of these six main types has its factors in a different 

 chromosome, and it was proposed, for convenience's sake, to call 

 these rods after them. 



The seventh one has been called by us the central one. 



It contains the factors, which differentiate the two constituent 

 parts of Oenothera Lamar ckiana, viz. laeta and velutina, from one 

 another, and contains moreover the majority of the atypical muta- 

 tions, which behave in the same way as those of other organisms 

 and might therefore well be designated as Drosophila-Yike. Like the 

 other chromosomes it has one typical mutation, but this is a very 

 rare one and has only been discovered in recent years. It has been 

 called 0. pulla. 



Besides the seven typical mutants there are two others which 

 constitute old and permanent races of almost the same rank. They 

 have been described, since the beginning of my study, under the 

 names of 0. mut. albida and 0. mut. oblonga. This brings the total 

 number up to nine, and it is evident that at least two of them must 

 be supposed to have their chromomeres in the same chromosome 

 with another linkage group. Thus there must be some chromosomes 

 with one and others with two, and perhaps even some with three 

 or more mutable linkage groups. It may be expected that such 

 allied classes will show some kind of linkage between them. As a 

 matter of fact these linkages are among the oldest instances known. 

 0. albida is produced from 0. lata in almost every culture and 

 every generation and in relatively high proportions. The same is 

 true for 0. oblonga in respect to 0. scintillans. The coefficient of 

 mutation is for most of the mutants, as a rule, less than 1 %, and 

 lies ordinarily between 0.05 and 0.7% and this among the seeds 

 of the pure 0. Lamarckiana as well as among those of 0. lata, 0. 

 scintillans and other mutant races. But 0. albida has been reported 

 to be produced from 0. lata in about 2-4% and 0. oblonga from 

 0. scintillans in 3-25% (Mutations-Theorie, I, p. 172-173, p. 249, 

 and Gruppenweise Artbildung, p. 257—258). These facts indicate 

 that the chromomere for 0. albida is a part of the lata rod and that 

 for 0. oblonga a part of the scintillans chromosome. From other 

 considerations it has been made probable that these two chromo- 

 somes are large ones, and it is of course easy to understand that 

 exactly the larger ones should have a double constitution. 



During a series of years I have counted the rosettes of albida 



