MR. O. BENTHAM OF EUPHORBIACEiE. 187 



to save the need of successively reading through on every occasion 

 a number of detailed descriptions before you could determine a 

 plant sufficiently to study it. And this defect was not remedied 

 by the short heading prefixed to each genus, stating that it is 

 some other genus with a difference. It is true that in some 

 cases species have been generically separated from a previously 

 known genus on account of some one specially exceptional cha- 

 racter, and that it is useful to call attention to the fact that such 

 is the only difference ; but it is also true that most genera differ 

 from every other one in several more or less prominent charac- 

 ters and are often equally allied to two or more ; the selecting, 

 therefore, one of them for comparison and indicating one point of 

 difference, amounts practically to little more than a general pro- 

 position that every genus is any other one with a difference. On 

 the other hand, the plates illustrating some of the most important 

 characters of each genus are most useful, the analysis very accu- 

 rate and well designed, the execution of the figures all that could 

 be desired ; one can only regret that they should have been incon- 

 veniently crowded in the plates without order, so that it takes 

 occasionally some time to find out which of them belong to any 

 one genus. The preliminary observations, occupying more than 

 one third of the volume, contain much on structure and develop- 

 ment, as well as on affinities and geographical distribution, well 

 deserving of study, though one may perhaps not always agree 

 with the author's conclusions. Dr. Baillon followed up this work 

 by various papers in his ' Adansonia,' entering into specific details 

 as to the Euphorbiacese of various collections, especially on Afri- 

 can ones in the 1st and 2nd volumes (1860-G1), New-Caledonian 

 in the 2nd volume (1862), and American ones in the 4th and 5th 

 volumes (1863-65). 



In the meantime the advance of the ' Prodromus ' required an 

 immediate working up of the whole of the species of Euphor- 

 biaceae, amounting even then to nearly three thousand. M. De 

 Candolle was fortunate in obtaining the services of Edmond Bois- 

 sier for the vast genus Euphorbia itself, of which the head quar- 

 ters may be said to be within the region of which M. Boissier had 

 specially studied the flora. His excellent enumeration of above 

 six hundred species was published in the second part of the 

 fifteenth volume of the ' Prodromus ' in 1862, and was liberally 

 followed up by illustrations in 121 plates (published in 1866 under 

 the title of ' Icones Euphorbiarum '). The elaboration of the great 



