42 Transactions of the [Sess. 



ever, left it to be more fully developed by a Frenchman, Antoine 

 Laurent de Jussieu. Born at Lyons in 1748, and educated at 

 Paris under his uncle Bernard de Jussieu, he was, when compara- 

 tively young, appointed to the office of Demonstrator in the 

 Jardin des Plantes. Li attempting to classify the plants of the 

 garden according to the artificial classification then in use, he 

 became thoroughly dissatisfied with it, and was led to consider 

 what a true classification ought to be. When he discovered Ray's 

 definition of a classification, he entirely adopted it, and worked it 

 out to greater perfection than Ray himself had done. He found 

 that, though the true aim of a classification had been reached, 

 there was considerable diversity of opinion as to the means of 

 arriving at such a classification. Tlie aim of this classification 

 being to arrange the vegetable forms in allied species, their 

 principle was, " That species which are dissimilar should not be 

 brought close together, nor should species that are similar be kept 

 apart." This gave rise to the question, " What is to determine 

 similarity or dissimilarity ? " The advocates of the artificial 

 classification selected each a cei'tain organ arbitrarily, and classified 

 according to the similarity or dissimilarity of that organ. They 

 differed greatly as to the organ which they selected. Andreas 

 Cfesalpinus (1583) chose the character of the fruit ; Dr Morrison 

 of Aberdeen (1670), the character of flower and fruit; Tournefort 

 (1684-1700), the character of the corolla; and last of all, Linnteus 

 (1778), the character of stamens and pistils. Jussieu held that it 

 sliould be determined by natural distinctions alone, and was the 

 first to lay down the principle that affinity could only be deter- 

 mined by correspondence in structure. Tliis is a natural deduction 

 from the axiomatic truth, " If any two plants be alike in every 

 point in their structure, then these two are identical." Hence he 

 defined a species to be, " Individuals which are very much alike 

 in all their parts, retaining tlieir resemblances from generation 

 to generation." " Those species," he said, " are to be associated 

 which correspond in the greater number of their characters ; but 

 one constant is of more importance than several inconstant char- 

 acters." For example, the stamens and pistils are more constant 

 characters than either the calyx or corolla, either or both of which 

 may be absent, and hence of greater value as a basis of classifica- 

 tion. It was this fact that made the classification of Linnasus of 

 more permanent worth than any other artificial classification. 



But Jussieu, in order to establish a gradation, had to discuss the 

 relative values of the constant structures, — " the determination of 

 which," Lindley says, " is the pivot on which the operation of any 

 systematist must turn." He pointed out that the relative value of 

 any structure must depend upon the function whicli that structure 

 performs ; and that as the existence of a plant depends on its 



