OBSERVATIONS UPON DOUBLING OF FLOWERS. 379 



in the direction taken when placed under the favoring conditions 

 of culture. 



Garden plants illustrate with accelerated force the working of 

 the universal law of compensation. Fruits not only enlarge, hut 

 "become seedless. Therefore, the increase in the size and other 

 modifications in the flowers of such plants as are grown for their 

 blossoms is only in accordance with a general law. An augmen- 

 tation of floral parts is only a step beyond the increase in size of 

 parts already present, and may be largely a matter of convenience 

 in the arrangement of the parts in the bud. When we remember 

 that any augmentation in the petals, etc., would be seized upon by 

 the gardener, and if possible reproduced, the wonder is that the 

 increase is not greater than it is. It is not claimed that such an 

 augmentation is a direct advantage to the plant, any more than is 

 the exaggerated size of a cabbage-head or the thick, rich pulp of 

 a grape, especially when the cabbage splits open and falls apart 

 of its own weight, or the grape-pulp monopolizes the whole sub- 

 stance, and no seeds result. When the guiding hand of man is 

 withdrawn, cultivated plants soon or late find their way back to a 

 stable condition called the natural form, and are again able to 

 cope with their neighbors, depending entirely upon the conditions 

 attendant upon the wild state. 



The point that now calls for our attention is the development 

 of one floral organ out of another widely differing from it in 

 appearance. Augmentation, we have seen, is to be expected, but 

 metamorphosis usually brings surprise. The unnaturalness of 

 this arises in part from the constancy of organs in wild plants, and 

 the general impression that a manifest difference in structure and 

 use must indicate dissimilar origin of the parts. All of the vari- 

 ous organs of a flower are now, as before stated, generally consid- 

 ered as lateral outgrowths from the stem, and in a state of nature 

 their number, size, shape, color, etc., depend upon the service of 

 each in the economy of the plant. In origin and early growth, 

 therefore, there is no microscopic difference between the sepals, 

 petals, stamens, and pistils. As shown at the beginning of this 

 paper, by taking the whole range of wild plants, it is not difficult 

 to find all gradations, from the outermost sepal to the central pis- 

 til. If these various parts have a common origin — namely, in 

 minute cellular outgrowths afterward connected with the primary 

 axis by a vascular cord — the wonder is that each type is adhered 

 to so closely in the wild forms, and the surprise should be that 

 under the modifying conditions of culture more striking combi- 

 nations are not found. The petals (that is, the inner whorl of 

 floral envelopes) and the stamens (the outer circle of essential 

 organs) form the boundaries between the two primary divisions 

 of the complete flower. It is here that the line of separation is 



