224 LOCK : 



far crosses between thoroughly distinct species could be 

 used as a source of evidence bearing upon the problems of 

 heredity. This object may be said to have been in some 

 measure accomplished. It has been clearly shown that a patient 

 study of the more fertile crosses is likely to yield evidence of 

 considerable value. Such work will be both tedious and expen- 

 sive, since long series of drawings or photographs of F 2 plants 

 will be required for comparison in the case of characters other 

 than the very simplest. 



So far it has been possible to demonstrate the existence of 

 MendeHan proportions in the case of a few simple characters. 

 There is no evidence that Mendel's laws are transgressed by 

 any character of colour. A curious character of shape was 

 observed to behave in the same way, namely, the funnel-shaped 

 or inflated throat of the corolla tube. This point is one of 

 some interest, because the difference between the two types 

 is one which the earnest seeker after adaptations might 

 suppose to be specially suited for determining the preference 

 of two different kinds of insects. Such an one would suppose 

 that the difference between the two types of flowers had arisen 

 through the extinction of a series of intermediate forms of 

 lesser degrees of fitness. I am not aware of the existence of 

 any evidence in favour of this view. On the contrary, every- 

 thing points to the conclusion that an intermediate type 

 would be just as fit for existence as either of the extreme forms 

 observed. As a matter of fact the most frequent insect 

 visitor in Ceylon appeared to be a sphinx moth, which sipped 

 the honey without ever alighting upon the flowers ; yet the 

 ripe capsules were full of perfect seeds. 



On the other hand, the fact that in experiment one of the 

 types was observed to arise from the other at a single step — in 

 the case of funnel-shaped offspring of inflated hybrid parents — 

 may be regarded as evidence to show that it is at least jiossible 

 that the original differentiation may have taken place in a 

 similar discontinuous fashion. 



Leaving out of account, the simple characters already men- 

 tioned, the casual observer might conclude that chaos reigned 

 among the offspring of these crosses. But experience of natural 

 processes geneially, and of genetic research in particular, 



