310 Nutrition and Selection. 
follow rapidly on one another. But if the conditions are 
unfavorable, as in a room, differentiation proceeds more 
slowly. The internodes tend to become abnormally long, 
to produce too little wood, the leaves develop small pin- 
nules only, and in very unfavorable conditions I have 
sometimes observed an interruption in the series of leaf- 
forms on the stem. Above the lyrate leaves simple ones 
were again formed, the series turning backwards. 1 
These phenomena are much better illustrated in those 
cases in which the first leaves are more compound than 
the later ones; for instance in the species of Acacia which 
produces phyllodes in reference to which GOEBEL'S im- 
portant investigations have thrown so much light on the 
relation between embryonic forms and external condi- 
tions. 2 I have already referred to this above; but I 
might now mention a figure of a seedling of Acacia 
I'crticillata which, after it had already reached the stage 
of producing phyllodes, was induced to repeat the bi- 
pinnate form of the embryonic leaves by unfavorable 
conditions. In the same way the production of linear or 
arrow-shaped leaves of Sagittaria sagitti folia and that 
of the perforated leaves of Momtera deliciosa and others 
was shown to be dependent on external conditions. In- 
sufficient nutrition tends to bring about a recurrence of 
the embryonic form, and it seems to be a secondary 
question whether this is the simpler or the more com- 
plicated. The Campanula rotundifolia studied by GOE- 
BEL, the flowerstalk of which changed from the linear 
to the heart-shaped form of leaves, 3 is perhaps the best 
1 See also E. ROZE, La transmission des formes ancestralcs dans 
Ics vcgetaux, Journ. d. Bot, Annee X, Nos. I and 2, 1896. 
2 K. GOEBEL, Organogra>phie der Pftanzen, I, p. 150, Fig. 105. 
3 GOEBEL, Flora., 1896, Vol. LXII, Ft. I. 
