376 W. Batesoll. 



of the phalanges, and those indicated by the behaviour of 

 double flowers in some plants are perhaps the only such 

 examples yet known. A case deserving- further study is that 

 of Oxalis tetraphylla, which hybridised with an ordinär}- 

 triple-leaved species ilatifoUa) gave fertile hybrids, for the 

 most part triple -leaved (Hilde brand (59)), Graham 

 Bell's (11) observations on multi-nippled Sheep neggest that 

 this character may have a similar behaviour. 



5. Blue colour is genei'ally dominant over white in the case of 

 the seeds of Zea Matjs, but both Cor reus (26) and Lock (70) 

 have observed irregularities and some failure of dominance 

 in this case. 



6. Spelt, namely long ears, are usually dominant over short or 

 club-shaped ears in Wheat (Spillman, Biffen, Tschermak. 

 and others). But both Spillman (89) and Biffen (12) have 

 seen cases of a blend-form appearing as F^ in certain cases. 



7. In Campanula and Mimulus Correns(38) has shown that 

 petalody of the calyx (the hose-in-hose or Calycanthemous 

 form) is dominant over the normal type of flower, but there 

 is great variation in degree. 



8. In Guinea pigs the rough-haired or rosetted condition is 

 usually dominant over the smooth-haired or normal condition, 

 but irregularities occur, and in some cases smooth-haired 

 animals are capable of reproducing the rough-haired condition 

 (Castle (16)). The Angora-hair in this animal also exhibits 

 complexities in heredity and does not follow the simple rules 

 found in the case of the rabbit (Castle). 



9. In Barley the hoods [Kapuzen) are generally dominant over 

 the awns (Biffen (12)) but many irregularities occur. 



The question naturally arises whether these irregularities of be- 

 haviour do not merely indicate the need for further analysis, and it 

 can scarcely be doubted that "pure lines", as Johann sen has called 

 them, could be raised in each of these cases, which would have a 

 uniform and consistent behaviour. As will appear from a consi- 

 deration of the phenomena of colour-inheritance there may be a close 

 superficial resemblance between types which diifer entirely in their 

 heredity, and it can only be by a very strict analysis that the homo- 

 geneity of such races can be determined. The same suggestion 

 applies of course to the cases in which mixture of forms (Pleiotypie of 

 Tschermak) occurs in Fj. 



Special attention should be called to the fact that in certain 

 strictly Mendelian cases F^ is normally a blend-form. This has been 

 seen in the case of ear-leng-th in Wheat (see above. List IT No. 6) 



