320 Nutrition and Selection. 
latent condition. On completely latent characters no 
effect can be produced; we are dealing solely with half 
and intermediate races. 'In other specimens, however, 
this is obviously not the case," says GOEBEL, "they retain 
their normal form even when grown in rich soil ; the 
high nutrition operates on the malformation, not as a 
cause, but as a releasing factor." 1 
It is a familiar fact that many garden plants de- 
teriorate if they are allowed to remain for a long time 
in the same place. They exhaust the soil and must, 
therefore, be moved from time to time. This is true, 
for instance, of Pansies, Anemones, 2 Dahlias, Petunias, 3 
the crested forms of many ferns 4 and so on. MORREN 
planted out a specimen of Saxifraga decipiens which 
had hitherto borne normal flowers on stony ground, into 
good garden soil. In this it grew very vigorously, formed 
larger flowers than before, and manifested at first a 
slight transformation of its stamens into petals which, 
however, increased gradually during the course of the 
summer until ultimately the flowers became entirely 
double. In Hedychiwn coronarmm the structure of the 
flowers is also shown to be dependent on nutrition. 6 
Wild apples and medlars lose their thorns in a few years 
if they are transplanted to gardens, 7 and Carlina acaulis 
becomes the so-called Var. canlescens, in rich soil, a fact 
which has already been recorded by WOLFF in his Tlieo- 
1 K. GOEBEL, Or gano graphic, I, p. 159. Various instances are also 
given by BURKILL, Journ. Linn. Soc. Bot., Vol. XXXI, 1895, pp. 2i8ff. 
2 VILMORIN-ANDRIEUX, Les fleurs de pleine terre, p. 87. 
3 HILDEBRAND, Ber. d. d. hot. Ges., Vol. XIV, 1896, p. 327. 
4 LOWE, cited by GOEBEL, loc. cit. 
'"Bull. Acad. R. Belg., Vol. XVII, Pt. I, p. 424. 
6 FR. MULLER, Flora, 1889, Pt. Ill, pp. 348-352, PI. 16. 
7 DE CANFOLLE, Physiologic vcgctalc, II, p. 721. 
