&■ 



1885.] FORMS OF LEA VES. 335 



leaves is another form of jirotection ; on herbs the presence of 

 hairs is often associated with that of honey, as they protect the 

 plants from the visits of creeping insects. Hence perhaps the 

 tendency of water species to become glabrous, Pohjijonum ampM- 

 hium being a very interesting case, since it is hairy when growing 

 on land, and smooth when in water. Sir John then dealt with cases 

 in which one species mimics another, and exhibited a striking photo- 

 graph of a group of stinging nettles and dead nettles which were so 

 much alike as to be hardly distinguishable. No one can doubt 

 that the stinging nettle is protected by its poisonous hairs, and it is 

 equally clear that the innocuous dead nettle must profit by its 

 similarity to its dangerous neighbour. Other similar cases were 

 cited. 



He had already suggested one consideration which in certain 

 cases determined the width of leaves ; but there were others in 

 which it was due to other causes, one being the attiti^de of the leaf 

 itself. In many genera with broad and narrow-leaved species, 

 Drosera and Plantago, for instance, the broad leaves formed a 

 horizontal rosette, while the narrow ones were raised upwards. 

 Fleshy leaves were principally found in hot and dry countries, 

 where this peculiarity had the advantage of offering a smaller 

 surface, and therefore exposing the plant less to the loss of water 

 by evaporation. 



Sir John then passed to aquatic plants, many of which have two 

 kinds of leaves : one more or less rounded, which floats on the 

 surface, and others cut up into narrow filaments which remain below. 

 The latter thus present a greater extent of surface. In air, however, 

 such leaves would be unable to support even their own weight, 

 much less to resist any force such as that of the wind. In perfectly 

 still air, however, for the same reason, finely-divided leaves may be 

 an advantage ; while in comparatively exposed situations, more 

 compact leaves may be more suitable. It was pointed out that 

 finely-cut leaves are common among low herbs, and that some 

 families which among the low and herb-like species have such 

 leaves, in shrubby or ligneous ones have leaves more or less like 

 those of the laurel or beech. 



An interesting part of the subject is connected with the light 

 thrown by the leaves of seedlings. Thus the furze has at first 

 trifoliate leaves, which gradually pass into spines. This shows that 

 the furze is descended from ancestors which had trifoliate leaves, as 

 so many of its congeners have now. Similarly in some species, which 

 when mature have palmate leaves, those of tlie seedling are heart- 

 shaped. He thought that perhaps in all cases the palmate form was 

 deri\ed from the heart-shaped, and that when in any genus we find 



