154 CO-OPERATION AND THE ORIGIN OF FLOWERS. 
‘precociously at five. four, or even three years old. 
This is said to be a certain sign of ill-health.* 
g. A certain orchid, Gastrodia Elata, which is a rootless 
saprophyte, only flowers when attacked by the 
Honey-fungus Armillaria mellea.°° 
It would seem at first sight that all these heroic methods 
and observations are too hopelessly different to admit of any 
common explanation. All nine of them, however, involve a 
critical and dangerous time for the plant, which has to 
undergo an exceedingly unpleasant experience. 
Moreover, as regards Nos. 1, 2, 4, 5, and 6, it is known 
by direct experiment that treatment of this kind increases 
the intensity of respiration. The crisis in the life of the 
plant, the effort to get over the injury, produces an intense 
respiration, which then induces precocious and _ profuse 
flowering. 
Now let us return to our Cycadlike ancestor of all flowers 
growing in a dry climate and in strong sunshine, to which 
uncomfortable conditions it was quite unaccustomed. It 
would surely be subject to excessive respiration, and some at 
any rate of its petals would turn red. 
But, besides excessive light, there would be two other 
evils, of which the effect must next be considered, namely, 
the loss of water in transpiration and the ravages of the 
Jurassic insect. 
The obvious effect of strong transpiration, especially in 
dry, exposed, and windy situations, is a dwarfed, stunted con- 
dition. Whereas in tropical jungles certain Rotangss are said 
te grow to a length of 200 metres, one finds on exposed 
summits in Tibet and the Caucasus hardly a single plant 
which is one foot, and a very large proportion which are 
only about one inch, in height. Euphrasia Officinalis on the 
Caucasus may be from one-half to three-quarters of an inch 
tall, and yet a perfect plant with flowers and seeds.2! 
In this country dwarfing also varies precisely with ex- 
* My own observations only refer to Conifers. More detailed 
information is wanted, especially of deciduous trees (sixth series). 
