parent species occur will be the area in which the great- 
est hybridization might be expected. While this concept 
is of great interest, both to the botanist and to the plant 
breeder, it must be realized that secondary centers of 
diversity may occur, through other factors such as hy- 
bridization or topography. 
4. Distribution of closely related species. Within 
broad limits, this criterion is quite useful. Some of the 
squashes (Cucurbita), for example, have been thought to 
be natives of Asia, but there are no wild species of the 
genus in Asia, whereas there are many in America. In 
studying the relationships of species, cytogenetics may 
at times give evidence which is of great value, as will be 
shown to be the case with tobacco (Nicotiana). 
5. Archaeological evidence. If sufficiently extensive, 
the record of archaeology may be of very great value. 
With the interest in carbon 14 dating, which is possible 
wherever there remains much organic matter, and in- 
creased attention to plant materials, this approach will 
become of increasing importance. Paleobotanical evi- 
dence, which is scanty for the cultivated plants, may 
conveniently be included in this category. 
6. Historical data. Written accounts may often be of 
value in interpreting recent dispersals of cultivated 
plants, and there remains to be done a good deal of care- 
ful work along these lines. 
7. Linguistic. The interpretation of routes of dispersal 
through the study of plant names is of interest and of 
some value, but this is probably the least reliable type 
of evidence and must be used with great caution. 
As in other such cases, the most dependable studies 
are those that use all available data, from every source 
or aspect of the problem. 
One of the untouched fields in ethnobotany is the study 
of the development and diffusion of geographical races 
[ 118 ] 
