412 History of the Sexual Theory. [BOOKIII. 



of the stigma to be the female principle, for experiment had 

 taught him, that if a stigma exchanges the moisture from 

 another stigma for its own, and is then dusted with its own pollen, 

 no hybrid form is produced 1 . In any case Koelreuter had a 

 more correct idea of the nature of sexual fertilisation than any 

 of his predecessors, and it was one specially adapted to enable 

 his contemporaries to understand the results of experiments 

 in hybridisation, while the hybrids themselves supplied most 

 convincing arguments against the prevailing theory of evolution. 

 We have arrived at Koelreuter's most important performance, 

 the production of hybrids. Here was a case for skilful ex- 

 perimentation, not for microscopic observation, and here he 

 obtained results in which nothing afterwards required to be 

 changed, but which when combined with later observations 

 have been used for the discovery of general laws in hybrid- 

 isation. The first hybrid which he obtained by placing the 

 pollen of Nicotiana paniculata on the stigmas of N. rustica, 

 produced pollen that was impotent ; but he soon after obtained 

 hybrids from the two species which produced seeds capable 

 of germination, and in 1763 he described a considerable 

 number of hybrids in the genera Nicotiana, Kedmia, Dianthus, 

 Matthiola, Hyoscyamus, and others. In the last portion of 

 his great work (1766) he speaks of eighteen attempts to obtain 

 hybrids with five native species of Verbascum, and submits 

 Linnaeus' views on hybrid plants, which we have already 

 described, to a withering criticism. He shows at the same 

 time from experiment, that if the stigma of a plant receives 

 its own pollen and pollen from another plant at the same 

 time, the former only is effectual, and that this is one reason 

 why hybrids which can be raised artificially are not found in 

 nature. We must not attempt to give a detailed account of 

 his famous hybrids of the third, fourth, and fifth degrees, nor 

 of his experiments on other points, such as the reverting of 



1 See Gartner, ' UeberBastardzeugting ' (1849", p. 62. I have unfortunately 

 been unable to meet with the second continuation of Koelreuter's work. 



