FIRST LITERARY EFFORTS 199 



question of classification in general, showing that any classifi- 

 cation, however artificial, was better than none, and that 

 Linnaeus made a great advance when he substituted generic 

 and specific names for the short Latin descriptions of species 

 before used, and by classifying all known plants by means of 

 a few well-marked and easily observed characters. I then 

 showed how and why this classification was only occasionally, 

 and as it were accidentally, a natural one; that in a vast 

 number of cases it grouped together plants which were 

 essentially unlike each other ; and that for all purposes, except 

 the naming of species, it was both useless and inconvenient. 

 I then showed what the natural system of classification really 

 was, what it aimed at, and the much greater interest it gave 

 to the study of botany. I explained the principles on which 

 the various natural orders were founded, and showed how 

 often they gave us a clue to the properties of large groups 

 of species, and enabled us to detect real affinities under very 

 diverse external forms. 



I concluded by passing in review some of the best marked 

 orders as illustrating these various features. Although 

 crudely written and containing some errors, I still think it 

 would serve as a useful lecture to an audience generally 

 ignorant of the whole subject, such as the young mechanics 

 of a manufacturing town. Its chief interest to me now is, 

 that it shows my early bent towards classification, not the 

 highly elaborate type that seeks to divide and subdivide under 

 different headings with technical names, rendering the whole 

 scheme difficult to comprehend, and being in most cases a 

 hindrance rather than an aid to the learner, but a simple and 

 intelligible classification which recognizes and defines all great 

 natural groups, and does not needlessly multiply them on 

 account of minute technical differences. It has always 

 seemed to me that the natural orders of flowering plants 

 afford one of the best, if not the very best, example of such 

 a classification. 



It is this attraction to classification, not as a metaphysically 

 complete system, but as an aid to the comprehension of a 

 subject, which is, I think, one of the chief causes of the sue- 



