The Periodicity of Semi-Latent Characters. 327 
Space does not permit the extension of this list; 1 
moreover all I am concerned with here is to show that 
this scheme also holds good for the distribution of anom- 
alies on the plant. Unfortunately it often can not be 
exactly applied because it is complicated with the effect 
of external influences. Every shoot and system of 
branches has its susceptible period, during which the 
external factors which happen to prevail favor the pro- 
duction of the anomaly, or the type of the species, ac- 
cording to their nature. But apart from these numerous 
fluctuations the rule holds good where the material is 
sufficiently plentiful to justify the neglect of exceptions. 
In a culture of Specularia Speculum which I had in 
1892 I found the flowers to be partly tetramerous and 
partly pentamerous. The pentamerous ones occurred 
at the top of the stem and of the strongest branches of 
the second order, whilst all the remaining flowers were 
tetramerous. A similar difference between apical and 
lateral flowers is exhibited as a normal character by 
Ado.va Moscliatellina and by many other plants. The 
Finns sylvestris of this neighborhood often has its needles 
^ <j 
combined into groups of three instead of groups of two. 
Each such group of needles, as is well known, belongs 
to a single dwarf shoot or spur. I found that the tri- 
foliate spurs occurred almost exclusively on the stronger 
branches, and chiefly at their upper end; but where they 
did occur they were numerous and closely packed. As 
a rule it is the main stem itself which bears them, but 
sometimes the strongest branches do so as well. There 
occurred up to 15 trifoliate spurs on the same one year's 
growth of the stem, all close to the apex, mixed with 
1 On the question of periodicity in the branching of cereals see 
SCHRIBAUX in Journal d' Agriculture pratique, 1899, and RIMPAU in 
Landwirthsclt. Jalirbitcher, Vol. XXIX, p. 589. 
