M. Reifferscheid, in a paper communicated to the Roman Archæological Institute, has propounded the most natural and proper solution of the problem.[115] He observes that it is not necessary to go farther than the magazines of the neighbouring Emporium for an explanation of this immense mass of potsherds. Every kind of provisions brought to Rome in ancient times was stored in earthenware jars, not only wine, but corn, oil, and other articles of commerce. A fire, therefore, which consumed any part of the Emporium would leave rubbish composed in great part of fragments of earthen jars (dolia); and, since many such fires must have taken place in the course of ages, and immense quantities of earthen jars must have been broken in the process of unloading, it does not seem at all impossible that so large an accumulation of matter should have taken place. At Alexandria and at Cairo similar heaps of potsherds are to be seen outside the walls, and their extent, though less, as might be expected, than that at Rome, is such as to create astonishment in the traveller’s mind when he sees them for the first time. An attempt has been made by M. Reifferscheid to determine the earliest date at which we can suppose this gradual deposition of potsherds to have taken place, but the data upon which he builds his conclusion, that the accumulation forming the Monte Testaccio first began to be deposited in the time of the decay of the Empire, about the third century, are not by any means such as to produce conviction. [Sidenote: Tomb of Cestius.]