IV 
PREl’ACE. 
these were accompanied by a far greater number of analytical dissections 
of nearly every flower or fruit that came under my notice. This vast 
amount of original evidence stands as a justiflcation for the contents of 
this volume : it has all been collected and classified under the present 
arrangement and brought together in four volumes, which will be deposited 
ill the botanical department of the British Museum, so that it may be 
accessible to all who may be desirous of investigating the subject. These 
tracings have been obtained from plants now existing in the Hookeriaii 
Herbarium at Kew, in that of the British Museum, in the AVallichian 
Collection belonging to the Linnean Society, in the Herbarium of the 
Museum at Paris, in the private collections of Prof. De Candolle, of M. 
de Boissier, of M. Delessert, of the late Dr. Bindley, and my own Herba- 
rium I have also preserved in cartouches all the dissected parts of the 
very numerous flowers and seeds examined ; so that I have been able to 
look over the same again and again, with the object of ensuring the 
utmost possible accuracy. 
In the introduction, the general structure of the family is explained, as 
well as the peculiarities which clearly serve to separate it into distinct 
tribes, and to mark the limits of the 63 genera into which the order is 
here divided. The characters of each genus are amply detailed ; and the 
diflerential features which serve to distinguish each genus and each 
species from one another are generally stated. These details, including 
the descriptions of the species, were printed in succession, as before 
mentioned, at various intervals up to October 1867 ; but although the 
whole of the type of this volume was then in print, its issue could not be 
effected for want of the numerous plates required for its illustration. The 
execution of this was retarded by the prosecution of the researches termi- 
nating the second volume of this work and the completion of its 45 plates, 
all performed by my own hands, and not issued before the end of 1869. 
Since that time 1 have been occupied, in great measure, in preparing the 
67 plates for this volume. With the view of expediting this work, I 
