153 NOTICES OF BOOKS. 



other geological conditions may, by extending one islet in one direction 

 and the other in au opposite one, whilst their contiguous shores are de- 

 pressed, remove the two islets and their induced races to indefinitely 

 great distances; and that these changes may be accompanied with 

 others of climate that may alter the appearance of the species beyond 

 all probability of their common specific origin ever being even so far 

 assumed as to suggest experiments to prove them one. 



The admission of such facts (and who can deny them being both 

 philosophical in the abstract, and capable of absolute proof, to some 

 degree at least, under existing conditions ?) does appear to render the 

 attempt to arrive at any definite conclusion as to the limits of many 

 species a desperate one. All this M. de CandoUe feels and candidly 

 admits, and from the very horns of the dilemma he proposes that a 

 position may be taken up, upon the ground that permanence of form 

 amongst wild plants has been proved during the short period of our 

 experience, and upon the grand point that many existing species have 

 not changed since the days of the ancient Egyptians, or since the more 

 ancient period of the deposits of turf, etc.* 



That this fact however leads to no practical result, M. de CandoUe 

 admits, because it is impossible to ascertain the state of species during 

 many thousands of years; because of the uncertainty of the period to 

 which we are carried back, and because induced forms {formes de- 

 rivees — races) are probably less numerous than original specific forms. 

 The last point is regarded as very important, and the facts adduced 

 by M. de CandoUe as iUustrating it are extremely valuable. 



In the first place, he says that races produced by cultivation never 

 so far depart from their original form as to be mistaken for different 



t 



Nap 



All naturalists however will not go even so far as this ; they will deny that the 

 Egyptian monuments and reUcs, and the fragmentary remains of plants in peat bogs, 

 etc, aiford even tolerable evidence of specific identity. Let any unprejudiced na- 

 turalist examine the minute often solitary characters upon which so many species of 

 existing plants are founded, and then ask himself how it is possible to pronounce 

 two plants to be specifically identical without at least having flowers and fruit and 

 leaves of both ? Can any one doubt this, who will only take up two or three of the 

 best European Tloras, and see what differences of opinion there are as to what are 

 species and what not, amongst our commonest and biggest forest-trees,— oaks, elms, 

 pmes birches, etc. ? Even habit of growth is sometimes insisted on as in itself 

 proof of specific difference ! 



t This is assuming genera to be natural and not conventional groups of species. 



chara r^ ^^ cultivated forms of plants that afford, abstractedly, good generic 



