METHODS FOll PROVING ORIGIN OF SPECIES. 17 



native of China, as I have elsewhere shown. They called 

 the pomegranate, which had spread gradually from 

 garden to garden from Persia to Mauritania, the apple of 

 Carthage {Malum Funicum). Very ancient authors, 

 such as Herodotus and Berosius, are yet more liable to 

 error, in spite of their desire to be accurate. 



We shall see, when we speak of maize, that historical 

 documents which are complete forgeries may deceive us 

 about the origin of a species. It is curious, for it seems 

 to be no one's interest to lie about such agTicultural facts. 

 Fortunately, facts of botany and archaeology enable us to 

 detect errors of this nature. 



The principal difficulty, which commonly occurs in 

 the case of ancient historians, is to find the exact trans- 

 lation of the names of plants, which in their books 

 always bear the common names. I shall speak presently 

 of the value of these names and how the science of 

 language may be brought to bear on the questions with 

 which we are occupied, but I must first indicate those 

 historical notions which are most useful in the study of 

 cultivated plants. 



Agriculture came originally, at least so far as the 

 principal species are concerned, from three great regions, 

 in which certain plants grew, regions which had no com- 

 munication with each other. These are China, the south- 

 west of Asia (with Egypt), and intertropical America. 

 I do not mean to say that in Europe, in Africa, and 

 elsewhere savage tribes may not have cultivated a few 

 species locally, at an early epoch, as an addition to the 

 resources of hunting and fishing; but the great civiliza- 

 tions based upon agriculture began in the three regions 

 I have indicated. It is worthy of note that in the 

 old world acjiicultural communities established them- 

 selves along the banks of the rivers, whereas in America 

 they dwelt on the high lands of Mexico and Peru. This 

 may perhaps have been due to the original situation of 

 the plants suitable for cultivation, for the banks of the 

 Mississippi, of the Amazon, of the Oiinoco, are not more 

 unhealthy than those of the rivers of the old world. 



A few words about each of the three regions. 



