46 On the Production of Hybrid Vegetables. 



but I think that several of them would intermix ; perhaps the 

 purple-fruited Passiflora edulis, and the scarlet P. princeps, 

 with the hardy P. caerulea : this I intend to try next year. 



Many species of Pelargoniums are known to intermix 

 freely, and can scarcely be kept distinct in their generations. 

 I found no difficulty in blending even the simple-leaved P. 

 lanceolatum with P. citronodorum ; but I have failed in every 

 attempt to intermix them with either P. tricolor or P. zonale. 

 The beautiful mule known by the name of P. ignescens, 

 which derived its fine colour from the dust of P. fulgens, 

 appears to be generally sterile, but I am told that it has pro- 

 duced a few seeds, probably from the pollen of another 

 plant. P. ardens, though figured in Loddiges' Botanical 

 Cabinet* as a native of the Cape of Good Hope, is known 

 to have been produced in England from an intermixture of 

 P. lobatum and P. fulgens. In the same book Johnson's 

 mule Amaryllis Reginaj-vittata is also unaccountably repre- 

 sented -f- as a foreign species, under the name of A. spec- 

 tabilis. 



Much yet remains to be ascertained on the subject of hy- 

 brid intermixtures, and the propagation of new mules offers 

 an interesting and inexhaustible source of amusement. I 

 think that I could even make some of the natural species, by 

 attending to their affinities; for instance, I think I could 

 produce the curious Gladiolus quadrangularis (or abbreviatus) 

 by an intermixture of G. tristis and G. Cunonius ; and I have 

 produced a plant hardly distinguishable from G. versicolor, 

 by the union of G. tristis and G. hirsutus. 1 have a natural, 

 though I believe unrecorded, Australian species of Goodia, 

 * Vol. II. Plate 139. f Vol. II. Plate 159. 



