SUPPLEMENT 



The root-sto:k of the violet, Viola pedunculata, T. & G., is a very 

 deeply buried underground stem, being sometimes eight or ten inches 

 down. The color of this violet is not in conformity with the ortho- 

 dox coloration theory ; it seems to be better policy in this case for 

 the flower to make itself conspicuous than to appeals to the aesthetic 

 sense of the bees, at any rate this yellow violet succeeds better in Cali- 

 fornia than any of its blue congeners. In countries of greater rain- 

 fall, violets often produce what are called cleistogamic flowers, that is, 

 flowers that do not open at all, but fertilize themselves in the bud ; 

 but I have not succeeded in finding such flowers in our climate. In 

 spite of rather infrequent insect visits, violets mature abundant seed. 

 The seeds are usually flung from the capsule as the valves spring 

 apart. 



The shooting star, Dodecatheon, affords a sharp contrast to the 

 peony in its methods of pollination. Providing no honey and little 

 pollen, it must be showy to stand any chance of insect attention, and 

 a more striking combination of colors cannot well be devised. Its 

 device for self pollination at the last, is one practiced also by the 

 nightshade. I have rarely seen insects visit Dodecatheon Clevelandi 

 of the southern valleys, but the mountain species are rather frequently 

 visited by bees. It should be noted that the pedicels, at first erect, 

 recurve to allow the corolla to protect the pollen, but become erect 

 again in fruit. The capsules contain small seeds which mature 

 quickly. The seeds are distributed by what Kerner calls the 

 pepper-box method, and a very effective method it is. In some 

 species the valves of the capsule separate at the tip, leaving an open- 

 ing surrounded by minute teeth ; in others a small lid falls off, but in 

 both cases the small opening allows the seeds to escape only when 

 they are thrown out with considerable force by the swaying of the 

 capsules in the wind. The advantages of massing flowers in clusters 

 should be much emphasized. Small flowers scattered against a green 

 background are invisible to insects flying at any considerable height. 

 There is some suggestion here of the principle of cooperation, which 

 is more fully developed in Chapter XII on " social flowers." 



The calla is treated rather fully in the Reader. The difference in 

 temperature between the interior and the exterior air can be tested 

 with the thermometer ; it may be several degrees. The pistillate 

 flowers seem to me still receptive after the dehiscence of thestaminate 

 flowers on the same spadix, and callas do produce seed in our climate, 

 whether from close pollination or only from cross, I have not deter- 

 mined. Neither have I observed the Chinese lillies out of doors 

 sufficiently to speak of their pollination ; but their growth from the 



54 



