964 LIFE : OUTLINES OF GENERAL BIOLOGY 



new departures that count are small mutations which occur abun- 

 dantly even in "pure lines". Thus the secret of snapdragons seems 

 to confirm Darwinism. 



SUMMARY AS REGARDS VARIATION IN PLANTS 



While the fact of the origin of species by evolution is no longer 

 disputed, nor the operation of natural selection upon organic forms 

 any longer denied, the absence of any general theory or rationale of 

 variation in either the animal or the vegetable world is not only 

 generally admitted, but often regarded as inevitable or even hope- 

 less: variation to some writers being simply "spontaneous" or 

 "accidental"; to others, if not fortuitous, at least dependent upon 

 causes lying as yet wholly, and perhaps hopelessly, beyond our 

 present powers of analysis. 



A theory of variation must deal alike with the origin of specific 

 distinctions and with those vaster differences which characterise the 

 larger groups. To commence, then, with the latter, we may pose 

 such questions as: 



1. How comes an axis to be arrested to form a flower? 



2. How is the evolution of the forms of inflorescence to be 

 accounted for? 



3. How does perigyny or epigyny arise from hypogyny? 



4. How is the reduction of the oophore and differentiation of the 

 sporophore to be explained among cryptogams and phanerogams, 

 and why should the moss type be so aberrant and so comparatively 

 arrested ? 



5. How did angiosperms arise from gymnosperms? 



6. How did wind-fertiUsed flowers arise ? 



7. How are the forms of fungi, algae, etc., to be explained? 



Does the explanation of such questions really lie merely in the 

 operation of natural selection upon innumerable "accidental" 

 variations requiring separate explanation in every case, or is any 

 constant law of variation discoverable ? 



Let us note the parallelism of form exhibited in many of these 

 cases of unrelated organisms, and inquire whether this does not 

 give us some other clue to their origin. 



In phanerogams we find the raceme modified into the umbel 

 and the spike by arrest of the main axis or of the flower stalks 

 respectively. Suppression of both gives the capitulum, and, as 

 specialisation goes on, the convex flower-bearing surface of the 

 composite becomes flattened, as in Dorstenia, and finally deeply 

 hollowed, as in the fig. (see Fig. 162, 1-4). 



In simple flowers an indefinite number of modified leaves is 



