WHAT IS A SPECIES'? 939 



receive great help in his determinations from a study of the geographical 

 distribution of various forms. He must also pay attention to such characters 

 as enable the field naturalist to recognize various forms when living, in 

 nature, and not alone to those which are usually available in museum or 

 herbarium specimens. It would probably be difficult to find two species of 

 trees belonging to the same genus so distinct and so easily distinguished in the 

 forest as Terminalia arjuna, Bedd. and Terminalia tomentosa,W. & A.; yet sys- 

 tematists working with herbarium specimens find it difficult to separate them. 



Cultural experiments in the case cf trees require, of course, a considerable 

 time, but more attention should be paid to the kind and amount of variations 

 observable on one and the same individual tree which would aid the systema- 

 tist in his selection of specific characters. Attention should also be paid, 

 especially in the case of evergreen plants, to variations observable on one and 

 the same individual at different seasons. Many evergreen parasites, such as 

 Leranthus, when growing on deciduous species, for instance, would probably 

 show considerable seasonal variations. 



In experimental cultures particular attention would, of course, be paid to 

 what may be recognized as " doubtful specific characters." Many which are 

 usually included under the term habit may be so described, and which are 

 known to be dependent to a great extent on the conditions under which the 

 plants have developed, such as are, for instance, implied by the words annual, 

 biennial, perennial, herb, undershrub, shrub, tree, erect, prostrate, climber, 

 etc. Many curious undershrubs are known, e.g., species of Premna which are 

 separated chiefly on account of their habit from other species which are trees 

 or large shrubs. Again, many species are known to occur both as erect trees 

 or shrubs and also as climbers, but Butea superba, Roxb. is separated from 

 Butea f rondo sa, Roxb,., mainly on account of its climbing habit. 



