i9i6] DeVRIES— DIMORPHIC MUTANTS 277 



of rosettes. Of the remaining 54, about one-half or 25 have 

 flowered, the others reaching this stage approximately at the time 

 when they were pulled up and counted. One plant was a mutant, 

 being a metacUnous velutina, just as described in my book on 

 pp. 308-3 1 1 . The others were densa and laxa, as should be expected, 

 and agreeing with these types throughout their whole life. 



0. canaXO. Cockerelli. — A culture of 63 specimens embracing 

 4 cana, 5 laeta, and 15 velutina, which have flowered, and a large 

 number of rosettes of radical leaves. Two plants were mutants 

 of the type of 0. lata and one of them has flowered. Neither in 

 the rosette stage nor at the time of flowering have the plants of 

 the cana type showed any difference from ordinary 0. cana, the char- 

 acters of the father, also of its twin hybrid type, being invisible 

 in them. Such was the case in almost all the beds containing 

 the hybrids whose mother was cana, and this made the distin- 

 guishing and counting of this type quite easy and sharply defined, 

 and therefore fully reUable. Short narrow leaves of a gray color, 

 a slender spike with long, thin flower buds with nodding tips were 

 everywhere the same distinguishing marks. The laeta and velutina 

 had the ordinary type of these twins, as produced by 0. Cockerelli. 



The reciprocal cross yielded 19 annual' and 13 biennial laeta, 

 besides 4 annual and 23 biennial velutina of the same type. The 

 annual plants have flowered; the biennials became stout rosettes 

 in July and August. 



0. canaXO. Hookeri. — Represented by 25 flowering plants, 

 3 younger ones, and 40 rosettes, and among the flowering indi- 

 viduals 5 cana, 5 laeta, and 4 velutina. The cana were like those of 

 the pure type; the laeta and velutina did not differ from those 

 of the cross 0. LamarckianaXO. Hookeri, some of the velutina being 

 of a yellowish green in such a degree as not to be able to produce 

 a stem. The reciprocal cross yielded only two laeta, one of which 

 has flowered, among a culture of 60 specimens. The remainder 

 were velutina, 24 flowering plants and 34 rosettes of radical leaves. 

 The types were the same as those in the reciprocal cross. 



0. syrticolaXO. cana. — ^Represented by 60 specimens of the 

 tj^e of (0. syrticolaXLamarckiana) laeta and velutina. Of these 

 19 laeta and 6 velutina have flowered, reaching a height of 2 m. in 



