Tl PREFACE. 



indispensable for accuracy, and wliose adopted meaning could be 

 explained in the Work itself. 



In commencing this process, the Author originally conBidered that 

 a mere compilation might be sufficient. The British plants are so 

 weU known, they have been so repeatedly described with so much 

 detail, they are mostly so familiar to the Author himself, that it 

 appeared to him only necessary to select from published descriptions 

 the characters that suited his purpose. But he soon found that no 

 satisfactory progress could be made without a careful comparison 

 and verification of the characters upon the plants themselves ; and, 

 during five years that the present Work has been in hand, the whole 

 of the descriptions have been drawn up in the first instance from 

 British specimens (except in the few cases of doubtful natives). 

 They have been then compared with the characters given in 

 Hooker and Amott's 'British Flora' and Babington's ' Manual,' or 

 with detailed descriptions in some of our best local Floras. They 

 Jiave, in almost all cases, been verified upon Continental specimens 

 from various parts of the geographical range of each species ; and a 

 considerable number have been checked by the examination of 

 living specimens. The works of the best French, German, Swedish, 

 ItaUan, or other botanists have also been consulted wherever the 

 occasion required it. The dried specimens made use of have been 

 chiefly those of the rich collections at Kew, including the unrivalled 

 herbarium of Sir William Hooker ; but the Author has also availed 

 himself of numerous and repeated observations made during forty 

 years' herborizations in various parts of Europe. 



Supposing, however, that descriptions are so successfully drawn 

 up that the young botanist may readily identify them with the 

 corresponding plants, they alone are insufficient ; he cannot be 

 expected to read them all through tiU he comes to the one which 

 he is in search of. Some method of arrangement must be adopted. 

 They must be so classed as to enable him to refer, by as simple a 

 process as possible, to the identical description belonging to liis 

 plant. If he knows the name, and wishes to ascertain what kind 

 of plant it designates, an Alj)habetical Index is at once suggested. 

 For the converse problem, where the plant is given and its name 

 is sought for, some corresponding device must be resorted to, and 

 the more simple it is the better it will answer its purpose. 



The plans proposed and more or less adopted in botanical works 

 for a classification with this view, commonly called an Artificial 



