Page:Popular Science Monthly Volume 82.djvu/237

Rh on the continent before the arrival of Europeans. They are considered by botanists as one species. The wild relative teosinte has been thought to be not only a distinct species but a member of a different genus. There is good evidence, however, that there is not a much greater difference between teosinte and the maize nearest like it than there is between a number of the most distinct maize varieties. These facts make it reasonable to suppose that both types arose from a common ancestor slightly different from each.

Teosinte and maize belong to the tribe Maydeæ, a division of the Gramineæ or true grasses. Our final problem is to connect the steps in the evolution of maize that distinguish it from the more typical grasses and if possible to picture the restored original form. The data from which one can do this come from observations of thousands of crosses between the different maize varieties.

Sweet corn is probably the most recent type. Sweet corns are simply dent, flint, pop and floury types that have lost the ability to mature starch grains. This is proved by crossing it with starchy kinds. For example, dent corns crossed with certain sweet corns produce flint types in the second hybrid generation. Starchiness is put into the hybrid by the dent variety and the latent flintiness of the sweet variety appears.

In the same way crossing indicates that as the pop or poplike varieties increased in size by numerous slight variations, the flint, the dent, and the floury kinds were produced through the correlation between the structure of the seeds and their size. This brings us back to a many-branched pop-like variety, examples of which are common enough to-day.

Most maize varieties have naked seeds, a feature unlike other grasses including teosinte. The remaining members of the family have