SCALE-LEAVES, FOLIAGE-LEAVES, FLORAL-LEAVES. 635 



strands goes with a veiy peculiar form of leaf, which may best be compared to 

 an open fan. The Maiden-hair Tree (Ginlgo biloba, see fig. 150^) may serve 

 as an example of this distribution of the sti-ands, which, on the whole, is not 

 connnou. It is also observed in several true ferns (e.g. Adiantum arciuihun, 

 Acrostichum sphenophyllum, and Livingstonei). With regard to the Ginkgo, 

 it should be mentioned that as a rule only four distinct strands enter the 

 blade from the leaf-stalk ; two central which are very delicate, and two lateral 

 which are very strong, and from which arise a large number of fine, forking 

 strands running upwards. 



Besides the arrangements of strands in leaf-blades here described, there are 

 many which cannot readily be brought under the limits defined ; in the same 

 way there are intermediate forms which may be placed just as well in one as 

 in another of our artificial divisions, and which we trj^ to describe clearly by 

 connecting the technical terms together. For example, we find intermediate forms 

 between arched and reticular strands which are described as arched-reticulate, and 

 so forth. 



The fact should be emphasized that the distribution and arrangement of the 

 strands in any given species is remarkably constant. This, however, is by no means 

 the case in genera and families. Of course there are j^lant-families the whole of 

 whose members exhibit marked agreement in this respect, as, for example, the 

 RhinanthaceEe, Melastomaceje, and Myrtacese; but, on the other hand, instances are 

 not lacking where the reverse is the case. Thus the various genera of Primulacese 

 present the widest varieties, and even the individual species of the genus Primula, 

 difler more from each other in the arrangement and course of the strands than 

 perhaps the Myrtacese from the Bor'agineje. Nevertheless the accurate determina- 

 tion and description of the distribution of these strands in the leaves is very im- 

 portant for that branch of botanical study, the object of which is to provide criteria 

 for the discrimination of species. The careful investigation of the distribution of 

 strands in leaves is, perhaps, of the greatest interest to the palteo-botanist, the 

 investigator of pre-existing vegetation. Those parts of plants which have come 

 down to us from earlier periods, embedded in geological formations, consist prin- 

 cipalh' of single leaves or fragments of leaves, often of very insignificant appear- 

 ance. In these fragments often we cannot even recognize plainly the edge, much 

 less the whole contour of the blade; but the strands themselves, and the net- work 

 which intervenes between the coarser strands, may be distinguished on the smallest 

 fragments. Often enough the palseo-botanist has only such scanty remains to refer 

 to when he seeks information about the species of plants with which our globe was 

 covered in long-past ages. Consequently even the most insignificant-looking bit of 

 leaf -network becomes of the highest importance. Just as an investigator, busj- 

 with the history of the human race, can draw certain inferences from the characters 

 of a hardly decipherable papyrus roll about the state of the household, about the 

 political institutions, the customs, manners, and civilization of the population settled 

 more than two thousand years ago in the valley of the Nile, so can the botanist. 



