Ft'TCRE HOTANIST. 15 



tlie leaves aiv flat on the rock. foriniii<^- a rosette : tlie leaves will 

 also become more hairy and possil)ly a darker red; the Howers 

 will become more closely set together, very likely more numerous, 

 and perhaps smaller. If tliis plant and its neighbours in the glen 

 freely cross with one another, then a new species will not occur, 

 l>ecause any variation (except such as is immediately due to the 

 situation) will be stamped out by crossing with the original species. 

 If. however, this plant and others sown beside it carmot cross 

 with the ancestral form, these modifications may become liercdi- 

 tary, and in course of time a new species Avill arise. The i:)lant is 

 leally in a sort of island, and we know that in islands there are 

 often an enormous number of peculiar or endemic species, and it 

 is this absence of crossing with the parent species, combined with 

 changed conditions, which has produced them. There are three 

 ways, at least, in which this may act in our own country. 

 (1) The spot may be an island by position, so that crossing can 

 scarcely occur. Hence the importance of studying- localities. 

 The Ilieracium nitidum of Backhouse discovered by J. T. John- 

 stone in 1892 at Andrewswhinnie. could only, by an inconceivably 

 minute chance, be crossed with its parent. (2) It may be an 

 island through change in the ilowering period. If the plant on 

 the exposed spot blooms and finishes flowering before its relatives 

 in the glen begin to fiower it cannot be crossed with the parent. 

 Herice the importance of know-iug how- long a plant remains in 

 fiower, and when it begins and ceases blooming. (;]) It may be an 

 island through its insect visitors being different. It is obvious that 

 if the same insect does not occur in both places, crossing is im- 

 -sible. and hence the importance of insect visitors. 



I trust that in the preceding I have shown that the future 

 Hiitish botanist wall have plenty to do, and I give it as mj' 

 deliberate impression both that this study of the why and ]i(nv is 

 the most important of all botanical enquiries, and also that any 

 person who chooses can make the most valuable disco vei'ies by 

 careful observation in his own back garden. I could certainly 

 have expanded this paper to many times its present length, but I 

 forbear, trusting that some of these hints may induce others to 

 follow this fascinating- enquiiy. 



