Varietäten, Dcscendenz, Hybriden. 55 



more or less systematical group have "related" chromosome num- 

 bers: i. e. values which are all simple multiples of one and the 

 same cardinal number, and which enter into an arithmetical 

 Progression. 



An investigation of 11 species — comprising 5 genera — ofthe 

 family Chenopodiaceae showed that all had 6, 9 or 18 chromosomes, 

 whence the formula for Chenopodiaceae would appear to be x =: 

 3n(^2). 



Also in families which — like the Cotnpositae — are generally 

 considered to be highly unstable with regard to the chromosome 

 numbers, we find, on investigation of the separate subgroups, that 

 a certain regularity is apparent. The chromosome numbers for the 

 groups Heliantheae and Anthemideae can thus be characterized by 

 x=:8n and x = 9n respectively. 



In cases where it is desired to ascertain the possible relation- 

 ship of lesser systematical units, the chromosome number will at 

 times provide a last weighty argument pro or contra. 



Apogamy, and the occurrence of double or manifold chromo- 

 some numbers, must be regarded as generally due to hybridization, 

 whereby the chromosomes of the parent gametes are practically 

 added together - but are paired only incompletely, if at all — in 

 the Zygote. It must be presumed that when chromosome pairing 

 entirely falls to appear after hybrid fertilization, an indirect chro- 

 mosome binding (see p. 199) must take place — if the Zygote is to 

 be capable of development and propagation — whereby the number 

 of chromosomes is doubled. 



In Callitriche verna, the reduction division of the pollen mother 

 cells gives rise to a kind of heterochromosomes, which cannot, 

 however, be homologized with those found in the animal world. 

 Corresponding phenomena have previously been observed in other 

 plants, and especially hybrids. 



The mottled-leaved Japanese hop, Humulus Japonicus var. 

 alhoynaculata , afifords an instance of the fact that certain qualities — 

 here that of mottled colouring — are only transmitted to the offspring 

 through the sexual cells of the female plant — and especially 

 through the cytoplasm. 



Cell and nuclear division cannot proceed without an occasional 

 loss of material, e. g. parts of chromosomes, which must affect the 

 genotypic Constitution of the cell. 



It must be presumed that occasional hybridization can give rise 

 to the formation of apogamous, or at any rate sexually abnormal 

 "new species", and that these newly formed biotypes and their 

 offspring persist as special minor species of biotypes simply because 

 they are, from their Organization, excluded from fertilization, and 

 thus from the adjusting — or mutually supplementing — effect 

 which is, in the writer's opinion, produced by the alternation of 

 generations. 



Himiiihis lupulus can develope fruits containing an embryo on 

 pollination with various related plants. Pollination with Urtica ureiis 

 produced particularly large and strong fruits, though these were 

 incapable of germination. It is presumed that a truly heterogeneous 

 fertilization has taken place, and that the hybrid was capable of 

 developement as long as it was nourished by the mother plant, but 

 that lack of internal homogeneity rendered it altogether incapable 

 of independent activity. 



