classes was taken by the English invaders, intermarriage would seldom take place. The other is that there are abundant relics, among the English graves, of a long-headed race, which can fairly be identified with the Iberian type as modified by increasing civilisation; and but scanty relics of the broad-headed Celt.[39] This fits in very readily with the supposition that under the Celtic, and therefore under the Roman rule, the cultivating class was largely composed of the pre-Celtic race; and allows us to believe that the agricultural population was but little disturbed. But though the cultivators already at work were probably left as they were, it is very likely that they were joined by many new-comers. We can hardly suppose that _free_ English warriors would have settled down at once as tillers of the soil, toiling half the days of the week on land not their own. But Tacitus describes a class of persons among the Germans whom he repeatedly calls _slaves_, and speaks of as subject to the arbitrary authority of their masters. They were not, he expressly says, employed in gangs, as on a Roman villa; but each man had his own house and family, and rendered to his master no other service than the periodical payment of a certain quantity of corn, or cattle, or cloth. He goes so far as to compare this class with the Roman _coloni_, though they differed from them in not being _legally_ free. He calls our attention further to the presence of a number of _freedmen_, occupying a position but little above that of slaves. There is no reason at all to suppose that Tacitus regarded these slaves and freedmen as few in number. And if there were slaves and freedmen in the same position among the invading English, they would readily fall into the ranks of the servile cultivators.[40]