404 BOTANICAL GAZETTE [December 



generation, the special mark of the mutant-father, and so are the same 

 in all the cases named. The mother plant may be O. biennis , O. 

 bienjtis cruciata (a form of O- biennis with linear petals which occurs 



in the dunes of Holland), O. muricata, or some other species of 

 Onagra, although not all of them are capable of producing twins. 



In the cases named the twins are widely different from their 

 reciprocal hybrids, but closely related between themselves. Their 

 characters often show transgressive fluctuability, and as a result of 

 this they are more easily distinguished at some periods of their Hfe 

 than at others. The better the culture and the stouter the plants, 

 the larger and more striking are the differences. Especially speci- 

 mens which contrive to make a large rosette of root leaves before they 

 send up their stem, and therefore begin flowering only late in the fall, 

 are hable to display their differential marks in a most striking manner. 



The twins are usually produced from the same cross in about 

 equal numbers. On account of their highly diminished fertihty it is 

 difficult to get large cultures of their progeny, for even when the pods 

 are large and contain an apparently full supply of seed, the germinat- 

 ing percentage is often very small. The following figures, however, 

 will suffice to state the fact. 



TABLE I 



O. LAETA AND O. VELUTINA IN EIRST GENERATION 



Number of Percentage Percentage 



Mother Father progeny of laeta of velutina 



O. biennis X O, brevistylis 85 47 53 



O. biennis X O. rubrinervis- 50 42 58 



O. b. cruciata X O. brevistylis 91 48 52 



O. muricata XO. Lamarckiana 58 61 39 



O. muricata X O. brevistylis 120 59 41 



O. muricata xO. nanella 59 58 42 



Average 52 48 



In this experiment the parents have been those described above, 

 and the crosses have been made by myself in the summer of 1905- 

 The countings have been made during the flowering- sea son of 1907, 

 only a small part of each of the cultures having remained rosettes; 

 but the distinguishing marks in these were as evident in the fall, 

 as those of the flowering plants. Some crosses, tried in previous 



me 



exact counting. 



t 



i 



