130 DELrillNIUM HENDEKSONII. 



the flowers gradually diminish in size, until, at last, they are scarcely more than one- 

 half the dimensions of the earliest ones. It follows, as a matter of necessity, that 

 the seeds ripened hy the flowers first expanded, will he much finer than those 

 produced by the latest blossoms ; and if, instead of permitting the whole of the 

 buds upon the spike, or raceme, to reach maturity, all but the first two or three are 

 pinched off, a still further improvement will result ; for the vital forces of the plant 

 will be concentrated in the development and maturation of a very small number of 

 blossoms. "When the plant produces several flower stems, one may be very well 

 spared for this purpose. It is not advisable, we think, to cut off the upper portion 

 of the spike ; we would simply pick off the buds, and that at the very earliest 

 period at which they can be removed, so that none of the strength of the plant be 

 wasted in nourishing them. The plants raised from seed thus obtained, may be 

 subjected to the same process ; and so on, for any number of generations, with 

 increasing benefit. It would, however, be necessary to guard against cross 

 impregnation by any other species of an inferior kind present in the same garden, 

 as this would tend to modify materially the character of the progeny. 



To return from this digression to the Delphiniums, which we must not quit 

 without a few remarks on the structure of the flowers. Like many other genera 

 of the Eanunculus tribe, the Larkspurs are remarkable for their coloured sepals, the 

 true petals being comparatively small. Of the five divisions of the calyx, the 

 upper one is prolonged into a hollow spur, which varies in length in the different 

 species. In the true Delphiniums, the petals are four in number ; two usually very 

 narrow, and furnished with spurs, which are included in that of the upper sepal ; 

 the other two are much broader at their ends, and reflexed downwards over the 

 stamens. In B. Hendersonii, and many other species, these two petals are covered 

 with yellowish hairs, about the centre, which communicates to the flower the 

 ap*pearance of being occupied by a Bee, or other large insect, which may be supposed 

 to have crawled in in quest of nectar ; and thence the popular designation of Bee 

 Larkspur, applied to D. datum, and other species. 



The distinct carpels of the perennial Larkspurs should be examined ; they are 

 very characteristic of the Order, and afford a good idea of the follicle, or single carpel, 

 opening by one suture. They are generally three in number, and contain one or 

 two rows of angular seeds. 



The annual species have been recently separated by Dr. Lindley from the genus 

 Delphinium, on the ground that the petals are united into one piece, as may be seen 

 in the common varieties. They are now, therefore, known as a distinct genus, 

 Consolida, in allusion to this union of the petals. Their single carpel is also a 

 feature of distinction not to be overlooked. 



Delphinium, from the supposed resemblance of the spur of the flower to a dolphin 

 (delphinw). 



