OF THE GENUS ULEX. 43 



We may sum up the various causes which appear to have a ten- 

 dency to modify the seedlings of the Furze plant as follows : — 



The reduction of leaf-surface and production of spines is 

 favoured by : — Drought ; shade of other plants ; cold winds and 

 soils ; hereditary tendency to production of spines (due to animal 

 selection). 



Increase in the leaf-surface and production of trifoliate leaves 

 is favoured by : — Moisture ; exposure to light ; warmth ; ancestral 

 tendency to produce trifoliate leaves. 



In the adult plant the conditions under which they are grown 

 may tend to modify the earlier leaves developed on the lateral 

 branches ; but those formed later, whatever may be the conditions, 

 always develop the spiny character. 



It is interesting to note that seedlings intermediate between 

 those which possess a large number of trifoliate leaves and those 

 which have linear leaves only, are constantly found in which the 

 struggle between those conditions which tend to produce trifoliate 

 leaves, and those which tend to form only linear leaves, is seen in 

 the fact that they show all sorts of stages in the reduction of tri- 

 foliate leaves to simple ones, with very few of the former, and 

 never quite reaching the latter. One such seedling, collected in 

 the shade, had the long internodes, characteristic of seedlings with 

 linear leaves ; but the lower thirty leaves were neither trifoliate 

 nor simple, but were in all stages of transformation of the former 

 into the latter. Such cases are not uncommon, though it is rare 

 to find such a striking example. 



We have already seen that all the young leaves and leaflets on 

 a seedling are curved upwards at the margin to form a furrow 

 extending down the middle of each leaf to the stem, or, in the 

 case of the lateral leaflets of a trifoliate leaf, to the median furrow 

 of the leaf. They are also curved downwards in a very decided 

 manner, becoming in many cases parallel to the stem. It appears 

 exactly as if this were an arrangement by which water is enabled 

 to run off the plant and prevent it getting wet. On trying some 

 experiments, however, to test this, by pouring water on to a seed- 

 ling from a height, it was found that very little water indeed 

 escaped this way, and that it was nearly all conducted, by means 

 of the furrows on the leaves, to the stem, and then delivered in a 



