110 FAMILIAR GAIiHEK FIOTJ'ERS. 



It is an interesting plant; with Howers of singular con- 

 struction. In their simplest form they are four-parted, the 

 petals boldly three-lobed, the stigma not less boldly four- 

 lobed agreealily with the numljer of sejials and petals, the 

 stamens eight in number, four of wlii/h are sterile, and 

 thus in this particular the fourfold arrangement is undis- 

 turbed. Dr. Lindley, in describing the family of Onagrads 

 to which the Clarkia Ijelongs, says the numlier 4- prevails 

 through every one of the floral organs. In the Circaja or 

 enchanter's nightshade the nundjer is halved, there bring 

 but two sepals, petals, &c., and in Lopezia, a near relative 

 of the last-named plant, the customary numljcr seems to 

 be still further interfered with, for that genus shows but 

 one stamen ; in reality, howe^•er, there are two stamens, 

 one of them perfect and l>earing i\\\ anthei', the other 

 sterile and in the form of a spoon-shaped petal. Although 

 the petals are iu general of a large size and in a high state of 

 development, vet there is a teudencv among the species to 

 lose them; I have seen an entire plant of Clurkin jjiilrliel/n 

 with every flower apetalous — that is, without petals. One 

 species, ('larlna Sin iiiifri , is alwa\-s so. 



Amateurs, generally speaking, are not sufficiently ac- 

 quainted with liaril\' annuals. Thev sow the seed in spring, 

 and in due time see the flowers, and jierliaps suiter a shadtiw 

 of disappointment. Or if quite satisfied, perhaps they have 

 no idea of the difference between the flowers the\' have and 

 the flowers they might have by a scientiiic mode of pro- 

 cedure. It may fie assumed that thev sow the seed late, so 

 that the jilants are hurried into flower before thev have had 

 time to make a deep root-hold. It may further l>e assumed 

 that they sow the seed too thickly, and do not thin out tlie 

 plants with sufficient courage, the result in this case being 



