Problem of Xeromorphy in Marsh Plants. 847 
There yet remains, however, one factor to be briefly considered, i. e. 
light. We have seen that in Spiraea a humid atmosphere has relatively 
little effect in preventing hair formation unless coupled with light of low 
intensity. Many observers have regarded light as the determining factor, 
especially in the formation of palisade tissue. In practice it is extremely 
difficult to satisfactorily control habitats in such a way as to distinguish 
clearly between the effects of light and relative humidity : two factors very 
closely associated in nature. Critical researches in this direction are much 
needed. In the absence of these it is impossible to say anything definite as 
to the mode of action of light. Failing evidence to the contrary, the writer 
somewhat inclines to the view of Vesque and Viet : 1 i. e. that light acts by 
influencing transpiration. Even in a still, humid atmosphere, intense light 
would certainly tend to increase transpiration, by reason of its heating 
effect. This view makes it less difficult to understand why the combined 
effect of two such apparently different factors should be greater than when 
either acts alone. At present, however, the possibility cannot be excluded 
that light may act either as a direct, independent stimulus, or that we may 
be dealing with a case of ‘associated stimuli \ 2 
So far the case has been made out most clearly for hair production in 
Spiraea Ulmaria. But in the light of our knowledge of the conditions 
which favour the formation of hairs (including root-hairs) and palisade cells 
in other plants (cf. §§ 6 and 9), it would seem not improbable that similar 
factors are at work in these also. In other words, that we are dealing with 
a more or less general phenomenon. 3 
An interesting apparent exception to what seems to be the general rule 
regarding the hairiness of successive leaves of a shoot is the case of certain 
South-west African plants recorded by Schinz. 4 In a number of species of 
Pseudobarleria , &c., the first-formed leaves are markedly hairy ; the later 
ones, however, are much less hairy. This is the converse of Spiraea. But 
the more hairy leaves are apparently unfolded towards the close of the hot, 
dry season ; the less hairy ones during the cooler period of rains. So that 
here, too, hair production shows a dependence on external factors similar to 
the cases described in this paper. 
4. Acting on the assumption that hair formation in Spiraea is affected 
by the amount of water supplied to the apical bud, some experiments were 
tried on growing shoots. When a number of leaves had been unfolded and 
were vigorously transpiring they were all cut off, subsequent leaves being 
also removed as soon as mature. It was thought that the removal of the 
transpiring surface, after a vigorous water current from the roots had been 
1 Vesque and Viet (’81), p. 175. 2 Cf. Darwin (’06). 
3 It is of course possible that in special cases the stimulus may be different, either in nature or 
degree : e. g. when palisade cells are formed in aquatic plants. 
4 Schinz (’94), pp. 67-70. 
