Primrose^ Oxtii\ Cowslij), and Polyanthis. 4^7'. 



complished naturalist possesses in so great perfection, and to 

 persuade us that it is not unlikely that this will, after all, prove 

 to be of little or no value towards the final solution of the 

 question. It should now seem that nothing but the multiplied 

 results of direct and accurate experiment can be allowed to 

 form the basis of our speculations in this, any more than in 

 every other department, of science. One well authenticated 

 fact will easily shake the most specious conjectures ; and, if we 

 do not listen to Nature when she is taking her own method of 

 opening to us her mysteries, we may be assured that we have 

 that spirit within which would rather be led by its own fancies 

 than bow even to the still higher authority by which Nature is 

 herself directed. Cultivation, we know, produces numerous 

 and very strange varieties of the same species; and, what 

 seems highly remarkable, these varieties, when once esta- 

 blished, fi'equently continue permanent through a succession 

 of crops raised from them by seed, except now and then, 

 where an accidental return to what is considered to be the 

 original stock takes place. Thus, to pass over the thousand 

 well known cases among our culinary plants, we find many 

 seedlings of the variegated sycamore striped like the parent 

 plant, though some of the same crop have lost this character. 

 But if it is no sure test, that a given form should cause an in- 

 dividual to rank as a distinct species, merely because we find 

 that form can be propagated unaltered by seed ; still less is it 

 any clue to the accurate discrimination of species, that the same 

 character should have been retained by the same individual 

 for many years. On the other hand, however, it is clear that 

 one single instance of change from one form to another, whe- 

 ther produced by seed or culture of the individual, if well 

 established, is quite sufficient to reduce any two of the most 

 permanent forms to the subordinate station of mutual varieties. 

 What has been hitherto recorded of the production of hy- 

 brids, has rather thrown confusion than order in the way of 

 this enquiry. The parents of many of these supposed hy- 

 brids may, after all, turn out to be no more than mere varie- 

 ties of the same, though, in some instances, they have, in all 

 probability, belonged to different species. What the law of 

 nature in this respect really is, we can hardly be said to have 

 ascertained. If we search for analogy in the neighbouring 

 study of entomology, we shall find that it is only very lately 

 that entomologists have ascertained that the nineteen varieties 

 of Coccinella variabilis, mentioned in Stephens's Catalogue as 

 found in Great Britain, form only a single species. Mr. Stephens 

 has recorded a similar reduction in the nineteen British varie- 

 ties of Coccinella mutabilis, each of which was formerly consi- 



