INTRODUCTION. 



15 



setum and Musci, in which the production of sexual 

 spores is as defective as is the production of pollen grains 

 in the hybrids of Aerogams. The obstacle to the regular 

 propagation of hybrids appears consequently to lie in the 

 development of those individual cells which have the 

 power to propagate the type of the parent form, and these 

 particular cells may or may not have the power of sexual 

 reproduction. At all events, more evidence must be 

 gathered before such a conception of a proposition of 

 such great biological importance is justifiable. As an 

 hypothesis this gives no explanation, but it may prepare 

 the way for the understandng of the conditions already 

 noted, since it unites under one heading a number of 

 different yet manifestly analogous phenomena in the 

 animal and vegetable kingdoms. 



FIFTH PROPOSITION. 



Malformation and odd forms, especially of the flowers, are in 

 hybrid plants much more common than in specimens of 

 plants of pure descent. As in Papaver, Dianthus, Pelar- 

 gonium, Nicotiana, Digitalis, double flowers also appear 

 to be produced with especial ease in hybrids. 



The Descendants of Hybrids. Hybrid plants are 

 more easily and more successfully fertilized by the pol- 

 len of the parent species than by their own pollen. Ex- 

 ceptions to this rule are rarely seen (as for instance in 

 Hieracium echioides X H . aurantiacum ) , but sufficient 

 experiments in this direction have not yet been made. 

 By their own pollen is understood the pollen of hybrids 

 resulting from the crossing of the same species, and not 

 only that of the identical specimens themselves. If hy- 

 brid plants grow in the neighborhood of their parent 

 species they must frequently be fertilized by these spe- 

 cies ; and in this case many intermediate forms between 

 the hybrid and the parents will appear in their progeny. 

 It has never been determined whether or not fertilization 

 of the parents could take place by the pollen of the hy- 

 brid. The common statement, that the progeny of a 

 hybrid are very variable, is therefore of but little value. 

 Occasionally also a hybrid is more easily fertilized by the 

 pollen of a third species than by its own as in Nicotiana 

 rustica X N. paniculata and Linaria purpurea X L. 

 genistcffolia. 



Progeny of Hybrids Fertilized by their Own Pollen. 

 (A X B) 9 X (A X B) 3 . (1) If fertile hybrids are 

 protected from pollenization by the parent plants or by 

 plants of a different species, one will obtain hybrid 

 plants of a second generation. It is my opinion that the 

 progeny of hybrids exhibit marked differences in the 

 duration of life. In long-lived plants the blending and 

 stronger union of the two types united in the hybrid is 

 frequently more complete, so that the progeny inherit 

 the characteristics of this new intermediate type. The 

 progeny of annual or biennial hybrid plants are, as a 

 rule, particularly variable and rich in different forms, as 

 in Pisum, Phaseolus, Lactu'ca,Tragopogon, Datura, Nico- 

 tiana alata X N. langsdorffii, and so forth. Exceptions 

 are found in Brassica, Oenothera, Nicotiana rust'ica X 

 N. paniculata, and Verbascum austriacum X V- nigrum. 

 The progeny of perennial plants behave in general in 

 a similar way, but the instances in which the interme- 

 diate type remains constant appear to be the more fre- 

 quent. Many of the hybrids often breed, moderately, 

 true, as in Aquilegia, Dianthus, La/vat era, Geum, Cereus, 

 Begonia, Cirsium, Hieracium, Primula, Linara, Veronica, 



Lamium, and Hippcastrum. The progeny of hybrid 

 shrubs and trees are in the majority of cases moderately 

 stable, as in JEsculus, Amygdalus, Prunus, Erica, Quer- 

 cus, and Salix; the progeny of many Fuschice and Cal- 

 ceolaria are constant. Some Rhododendron hybrids 

 breed true and a portion variably. The progeny of the 

 hybrids of Vitis, Pints, and Crdtcegus appear to be very 

 variable. 



2. The different forms in which many primary hy- 

 brids appear are usually not stable in their offspring. 

 In Dianthus the less-frequent forms ("Ausnahmetypen," 

 according to Gartner) usually revert to the normal hybrid 

 form. Mendel found that the different primary forms 

 of the Hieracium hybrids breed true. 



3. C. P. v. Gartner and other botanists have advanced 

 the proposition that the progeny of hybrids become 

 weaker and less fertile from generation to generation. 

 It is true that their vegetative power, which at first 

 is increased, is progressively decreased by self-fertiliza- 

 tion. Gartner's researches were, moreover, instituted on 

 a very small scale, so that not only very close inbreeding 

 but also the many circumstances which cause deteriora- 

 tion in garden-plants of which only a few specimens are 

 cultivated influenced his hybrids. Gartner himself no- 

 ticed exceptions in Aquilegia, Dianthus barbatus X D. 

 chinenss, and D. armeria. X D. deltoides. Hybrids of 

 nearly related species are often grown perenially with 

 ease, as in Brassica, Melandryum, Medicago, Petunia.' 

 Many gardeners assert with great positiveness that many 

 hybrids can be propagated by means of seeds through 

 many generations, as in Lychnis, Erica, Primula auricula 

 X P- hirsuta, and Datura.* Many observations on wild 

 plants seem to confirm these views. The theory has also 

 been advanced that the fertility of hybrids is increased 

 in later generations. It does not appear that such a rule 

 can have a universal validity. It is much nearer the 

 truth that many times fertile hybrids appear and that 

 they can easily increase under favorable environment 

 because of increased fertility. Fertile offspring of hy- 

 brids are, in fact, often products of back-crossings. 



4. Complete reversions to the parent forms without 

 influence of the parental pollen arise, except in rare in- 

 stances, only in hybrids of nearly related races. In such 

 hybrids true reversion appears only in a small number 

 of plants, as in Phaseolus. 



5. From the variable progeny of fertile hybrids sev- 

 eral dominant types are often produced in three to four 

 generations. If these new types are protected from 

 crossing they tend to become constant. Scientific re- 

 searches which confirm these statements have been carried 

 out in but small numbers, especially by Lecoq in Mira- 

 bilis, by Godron in Linaria and particularly in Datura. 

 Gardeners have produced many new races with well- 

 marked characteristics by crossing different species, and 

 many permanent wild intermediate forms have probably 

 originated in this way, as for example, Brassica, Lychnis, 

 Zinnia, Primula, Petunia, Nicotiana commutata, Pent- 

 stemon, Mentha, and Lamium. The new types of hybrid 

 progeny depart frequently in individual properties from 



* "Botanists say that species so produced" (i. e., hybrids) "revert 

 to either of their parents in the third or fourth generation, or 

 become sterile altogether. This is plausible enough in theory, in 

 the closet, but will not do in the potting bench." Beaton, quoted 

 by Loudon, Arbeit II, p. 944. 



