556 Prof. G. Klebs. Alterations in Developrneiit and [June 25, 



by environment to form rosettes in place of flowers continue to live during 

 the winter under very careful treatment. The stem becomes stronger and 

 thicker, the rosettes go on growing, we get peculiar forms (for instance, a 

 very diminutive tree, which also can live until the following year, and 

 perhaps even longer). Such plants come to flower after two or three years 

 and form a new inflorescence on the old stem. 



The flowers, the essential parts of the inflorescence, represent organs with 

 a very complicated structure, which is generally so constant for every species 

 that it can be used as the most important factor in the determination and 

 classification of the plants. However, this constant behaviour will be 

 observed only under the ordinary conditions of life. As soon as we apply, 

 at a suitable moment, special combinations of external influences, the 

 flowers are subject to a variability even more intense than that of the 

 vegetative organs themselves. 



The production of normal flowers depends, as we know, upon a certain 

 retrenchment of nutritive salts and on an increase of the organic substances ; 

 conversely, it is prevented by rich manure and moisture or by a decrease 

 of carbon assimilation. When we produce the conditions so favourable 

 to vegetative growth, either just before or during the first beginnings of 

 flowers, the flowers become altered in every respect and in a high degree. 



The best method is to let a well-fed rosette develop into an inflorescence 

 and then to cut off its top at the time of flowering. We further cultivate 

 such plants in soil rich in manure and moisture, and in combination with a 

 high temperature or red or blue light. 



In the leaf axils of the reduced inflorescence new flowers take birth under 

 strongly altered conditions and then offer a surprising richness of variation. 

 I have not the time to enter into all the details, but will only deal with 

 some of the more important points. The flower of Sempervivum Fnnkii has 

 a very regular structure, consisting of the same number of sepals, petals, and 

 carpels, and of twice the number of stamens. In the altered flowers the 

 number of all tliese parts, varying independently of one another, shows 

 strongly marked deviations. I'erliaps the various modifications of form in 

 all the members of a flower are of even greater interest. We observe the 

 transformation of sepals into petals, of petals into stamens, of stamens into 

 petals, of stamens into carpels, and of carpels into stamens; in short, wo 

 find most of the deviations long known as teratologicul cases in the garden 

 plants. But with tSem])crvivum the majority of the deviations had not 

 previou.sly been seen, neither under natural conditions nor under cultivation. 

 'I'lioy were first brought to light tlirougli tlie influence of external factors. 



Among sucli deviations tlic formation of doulile flowers by tlie traus- 



