BOTANICAL SUBJECTS. 37 



pairs of side leaflets have fused with the odd one, and formed a 

 palmate leaf.* 



One can fancy that a seed, ha\'ing been transported to a new 

 region, and producing a plant that could continue to live under 

 those circumstances, would often produce as many variations as a 

 wild plant often does, when brought under cultivation. Then a 

 struggle between all the variations would eliminate the weakest, 

 and select the fittest. 



If one keeps his eyes open, when looking at the flowers on the 

 same plant, it is almost next to impossible not to meet with striking 

 variations. 



Among a mass of honeysuckle flowers, I found that the 

 majority had four petals curled upwards, and either all fused 

 together, excepting their tips, or the two middle ones fused and 

 the two adjacent ones fused to only half their length, the fifth petal 

 being curled downwards as a lip. A certain number had two 

 petals curled downwards and separate, while three were curled 

 upwards and fused. More rarely there are two lower and four 

 upper, thus introducing an extra petal. More rarely still, all five 

 are curled upwards and fused, up to their middle. One with 

 six petals had also six stamens alternating with them. Whe 

 six petals occurred there were six stamens. One, however, had 

 seven petals, six up and one down, and only six stamens. 



If we turn to the cultivated Gladioli^ we often find on the same 

 spike some fiowers that have two outer petals below and one above, 

 or vice versa \ that is, one outer petal below and two above. f 

 Curiously enough, it is always the three lower petals, which are 

 distinctly marked, whatever the form may be, and the stamens and 

 pistil are always arched under the upper petal or petals. 



In the Hippeastrum we have a Gladiolus flower with its petals 

 reversed, that is, the distinctly marked ones are upwards, while 

 the bundle of stamens and pistil reposes on the lower petals. I 

 have seen a reversed Hippeastrum^ but the stamens, and this is 

 the extraordinary part, reposed on the lower petals, which in this 

 abnormal case were those distinctly marked. 



All these variations, often occurring on the same branch of the 

 same tree, may, under certain circumstances, and through the seed, 



* The leaf of Geum, may have easily resulted from the leaf of Potentilla. 

 t In another place I have endeavoured to account for this puzzle in the 

 Gladiolus. 



