278 Ostenfeld. 



In order to get better information, we must touch the great 

 problem of the importance of h3?bridization for the origin of species 

 in general and not restrict ourselves to a study of the papers on 

 Hieracia, but we will not lose sight of the fact that our principal 

 object is to answer the question with regard to the Hieracia. Generally 

 taken, the question runs: Is hybridization of any importance 

 as a species-producing factor in the Flowering Plants? 



But before we can try to answer this question, we must briefly 

 make clear what inheritance in hybrids means; in other words, we 

 must learn what, at the present time, the workers in the problem 

 of heredity say about the matter. 



In his "Mutationstheorie" (vol. II, 1902 — 03) and later in his 

 lectures (1906), Hugo de Vries has copiously treated the problem. 

 He distinguishes sharply between variety hybrids and species hybrids, 

 just by reason of the supposed difference in inheritance ("balanced and 

 unbalanced crosses"). In variety hybrids we find Mendelian segre- 

 gation in force, while "es eine nicht unerhebliche Reihe von Constanten 

 Rassen giebt, welche durch künstliche Verbindung von zwei ver- 

 schiedenen Arten entstanden sind, und sich im Laufe der Generationen 

 in jeder Beziehung, höchstens mit Ausnahme der verminderten Frucht- 

 barkeit, wie gewöhnliche Arten verhalten" (Mutationstheorie, II, 

 p. 73). He has himself produced a constant hybrid, Oenothera mu- 

 ricatax biennis, which indeed has onlj^ a very limited fertility; moreover, 

 he mentions as examples the hybrid Aegilops speltifonnis, famous in 

 the last century, and a hybrid Anemone silvcstris x magcllanica pro- 

 duced by JANCZEWSKI, and calls attention to A. Kerner's merits 

 for having emphasized the hybridization as an important factor in the 

 origin of species in nature. 



To be sure, the examples of natural hybrids, put forth by Kerner. 

 cannot be said to be unassailable proofs, as we do not know the 

 origin of these hybrids exactly, nor do we know whether the hybrids 

 now existing are several generations old (consequently constant) or 

 only the first generation which in course of time has propagated 

 vegetatively; then all the examples are perennial plants. Nevertheless 

 DE Vries seems to agree with Kerner's supposition and to be in- 

 clined to consider species hybrids as important and as constant. It 

 must not be forgotten, however, that by segregation through several 

 generations new constant forms can also be produced, and that in 

 the long run it is of no consequence whether the hybrid is constant 



