FLORAL BIOLOGY. 207 



all these and showed that in most cases there was a gain in 

 all three, but that it was usually very different in each ; a 

 large increase in one is generally accompanied by a small 

 increase, or even a decrease, in one or both of the others. 

 If we represent the value of a in crossed plants by a„ in 

 self-fertilised plants by a s , we find, for example, in Dianthus 

 caryophyllus — 



a x _ 100 b x 100 c x 100 

 a s 92 ' b 86 ' c s 65 ' 

 whilst on the other hand, in Nicotiana Tabacmn the ratios 



100 100 IOO / 1 . 1 \ c 



are , — -, (second generation in each case), bo 



150 178 97 



that, in considering the advantage (if any) of cross-fertilisa- 

 tion in any given case, each factor must be treated separately, 

 remembering at the same time that a certain (numerical) 

 gain in a may mean more to a certain plant than an equal 

 gain in b or c, and so on. We must remember further that 

 the plant lias to pay a certain pi'ice for its cross-fertihsa- 

 twn, viz., the .material and energy represented by large 

 corollas, honey, scent, extra pollen, big stigmas (in wind- 

 fertilised flowers), etc. The price to be paid will vary 

 considerably, even in the same species, being affected by 

 the climate, the competition of other flowers, the number 

 and nature of the insects in the locality (these varying from 

 place to place, and from month to month of the year), and 

 other factors. Again, cross-fertilisation means delayed fer- 

 tilisation, and in the short summer of arctic or alpine climates 

 where there is but little time for the ripening of seed, this 

 may become an important point. 



Similarly there will be variation in the gain to be 

 derived from cross-fertilisation in any given case. The 

 factors a, b, c, will enter into it in different degrees and 

 sometimes one may be more advantageous than another. 

 We know too that the fertility of a plant varies from one 

 country to another, so that the gain derived will vary too. 

 A further advantage probably lies, to many plants, in the 

 increased variability resulting from cross-fertilisation. It is 

 thus evident that in dealing with the natural history of a 

 flower we have a very complex problem to solve, and must 



