10 



NATURAL HISTOIIY OF PLANTS. 



stamens assume the appearance of organs supeq^osed in vertical rows,' 

 wliile eacli row is composed of a flir larger number of stamens' than 

 in X. Gar'uh'Ua. The gynscceum is composed of from four to five carpels, 

 which do not correspond exactly to either the sepals or their intervals. 

 Tliey are inserted very obliquely by their bases on the tapering 

 receptacle, so as to s^ive them the appearance of being united into a 

 many-celled ovary up to a certain height. Free above, they each 

 taper upwards into a style, furnished towards the tip with stigmatic 

 pa])ina'. Along the inner angle of each ovary is a multiovulate 

 phicenta as in N. Garidella, and the fruit consists of five follicles 

 united below and dehiscing by the inner angle. 



While all the other Nlcjellm have blue or whitish flowers, N. 

 onentaUa L., which has been made the type of a small separate 



group,^ has them yellowish, 

 with stamens closer together, 

 retaining the appearance of a 

 spiral arrangement, though 

 not divided into groups. The 

 carpels are very variable in 

 number (figs. 22 and 23), and 

 are only united by the lower 

 portion of their ovaries. The 

 seeds are flattened, orbicular, 

 and bordered by a narrow 

 membrane. 

 N. damasccBna L., which is cultivated in our gardens, lias been 

 considered the type of a special genus under the name of Erobatos,'' 



' " After the expansion of tbe tlower," says 

 P.\YEK {Organoy., 2 18), "the staminal radii seem 

 alternate with the petals ; but this is only in ap- 

 pearance, as they are truly superposed when 

 young." These stamens arise successively in a 

 spiral order, the same as the petals or nectaries, 

 which only confirms us in our opinion that they 

 are organs of the same nature. The anthei's are 

 introrse, and open like those of the Columbines. 

 But the line of dehiscence is not in the middle of 

 the cell, and the exterior wall, which is turned 

 back outwards is much broader than the interior. 

 This is the only dillerence. 



■■'In our gardens the flowers often become 

 more or less double ; all the stamens, or a limited 

 number of them beginning below, are trans- 

 formed into pctaloid scales, as appears to be 

 normal with the eiglit or ten lowest stamens. 



^ NigeUastrum McENCH, Meth., 311, 313. — 

 Spach, Suit, a Bnjf., vii. 310. The number of 

 carpels varies from five to ten and even more in 

 iV'. orieiitalis. The ovaries are narrow, com- 

 pressed against one another, and each surmounted 

 by a tapering, erect, straight style. In N. 

 corniculata DC, which belongs to the same 

 section, they are bent outwards. The carpels 

 may be only two or three in number in this 

 species, as in Garidella. 



* See Spach, Suit, a Buff., vii. 301. Ero- 

 hatos is there considered as a distinct genus, 

 while De Candolle only makes it a section of 

 the genus Nigella. N. coarctata Gmel., 

 which we have seen cultivated, does not appear 

 to us to differ specifically from N. damas- 

 cfpna. 



