358 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



unknown which are often inheritable. The experiments of Mons. Ber- 

 nard show us that by changing the fungi of orchids one may create 

 monstrosities ; that is to say, in short, transform the orchid. It may 

 be that the change thus brought about will be so profound that the 

 creatures thus formed may be unmanageable; but if they can be 

 brought to flower and reproduced, in all probability there will be some 

 very curious creations. It can be seen that breeders still have a vast 

 field to exploit. , 



As a sequel to these remarks, some deductions may be drawn. At 

 present too much attention is given to the study of hybrids. The 

 Mendelian laws, so long forgotten, and recently brought to light 

 again by the work of De Vries, Tschermak, Correns, and Bateson, 

 would lead one to believe that the key to the riddle of evolution had 

 been found. These laws, it must be stated, are applicable only to 

 very simple cases, such as that of two varieties which differ from each 

 other by one or a small number of characters. The characters of their 

 offspring are then subject to indisputable mathematical laws. They 

 do not seem applicable, at least at the present moment, to cases of 

 two parent species of an offspring, differing from each other by 

 numerous characters. If even these complex cases could be cleared 

 up and reconciled with Mendelian principles, the result would be a 

 theory that evolution takes place only in the ovule. Can we admit 

 that an exterior influence can never cause the appearance of new 

 characters? Upon this there can be no division of opinion. All that 

 has been set forth above with regard to the orchids pleads a contrary 

 case, which is in accordance with the theory set forth by Lamarck, 

 the famous disciple of Buffon. 



