938 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, Vol. XVII. 



correct. Errors which have been committed can then be subsequently correct- 

 ed, and this will ensure our classifications being built on a sound and 

 reliable foundation. In the case of plants, for instance, such proof 

 will be obtained from observations of the individuals raised from 

 seeds or brood bodies (under which term may be included all reproduc- 

 tive bodies arising asexually), these having been gathered from plants 

 which have developed wild in nature. In accordance with the suggestions 

 now made, these observations will be confined to such individuals, i.e., to the 

 immediate offspring of plants growing wild. The posibility of any character 

 which may be regarded as fixed and constant in nature being affected by the 

 artificial conditions of existence under cultivation is thus avoided. Moreover, 

 any factor which, in nature, is capable of producing an hereditable variation, 

 i.e., pollination by neighbouring plants, will, of course, have taken effect before 

 the seeds or brood bodies are matured, and the result will become manifest in 

 the young plants produced from such bodies. As only a brief period will thus 

 be required for each experimental culture, a large number of experiments can 

 be carried through in a short period and results quickly obtained. 



Some doubtful points may indeed be settled by a very few experiments. 

 Thus in a case where two groups of forms have been classified as distinct 

 species — if individuals of one group are commonly found in considerable 

 numbers among the immediate offspring of individuals of the other group, the 

 two forms are at once shown to belong to one and the same species. 



Other points, such as to whether forms can be considered sufficiently constant 

 to be ranked as undoubted species, i.e., whether their distinguishing characters 

 are fixed and .constant under varying conditions of existence, require of course 

 extensive series of observations. In such cases attention must be paid 

 to the conditions under which the parent plants are existing in nature, such 

 as, for instance, the presence of other plants by which they may have 

 been fertilized, and hence the reproductive bodies must be gathered 

 from individuals living in different localities. Then, again, the effect on 

 the young plants of temperature, light, available moisture, soil, air currents and 

 other factors which are known to affect the development and form of plants 

 must be noted, and it must not be forgotten that the effect produced by any 

 factor is probably dependent to a great extent on the particular stage of the 

 plant's life-history during which it acts, the plant being more susceptible at 

 some times than at others. Another important point to be remembered when 

 studying the effect of various factors is that no real distinction exists between 

 considerable variations occurring suddenly and slight variations which may be 

 gradually accumulated and fixed. A high temperature, for instance, acting 

 for a short time may cause a change which is only produced after a long 

 period by a lower temperature. 



11. Finally, it may not be superfluous to add that, for successful classifi- 

 cition, the naturalist in the field must, as far as possible, work in combination 

 ivith the naturalist in the museum or herbarium. The latter, for instance, 



