And, first, we must be quite clear that in those early ages there was nothing in Rome which we can call a temple, as we understand the word; nor was there any such representation of a deity as we can call an image or _eidolon_. The deities were settled in particular spots of ground, which were made _loca sacra_, _i.e._ handed over to the deity by the process of _consecratio_ authorised by the _ius divinum_.[291] It was matter of no moment what might be erected on this bit of ground; there might be a rude house like that of Vesta, round in shape like the oldest Italian huts; there might be a gateway like that of Janus; or the spot might be a grove, or a clearing within it (_lucus_), as in the case of Robigus or the Dea Dia of the Arval Brethren. All such places might be called by the general name _fanum_; and as a rule no doubt each _fanum_ contained a _sacellum_, _i.e._ a small enclosure without a roof, containing a little altar (_ara_). These "altars" may at first have been nothing more than temporary erections of turf and sods; permanent stone altars were probably a later development. Servius tells us that in later times it was the custom to place a sod (_caespes_) on the top of such a stone altar, which must be one of the many survivals in cult of the usages of a simpler age.[292] With such spots as these we cannot associate anything in the nature of an image of the deity established there; and we have every reason to believe that no such thing was known at Rome until the Etruscan temple of the Capitoline trias was built near the end of the regal period. Varro expressly declared that the Romans remained for more than 170 years without any images of their gods, and added that those who first introduced such images "civitatibus suis et metum dempsisse et errorem addidisse."[293] What he had in his mind is clear; he had indeed no direct knowledge of those early times, but he is thinking of a definite traditional date in the kingly period--the last year of the reign of Tarquinius Priscus, who, according to Varro's own account, built the temple on the Capitol and placed in it a statue of Jupiter.[294] That was the oldest image of which he knew anything; and, as Wissowa has remarked, his belief is entirely corroborated by the fact that in every single case in which the image of a god has any part in his cult, it is always either this Capitoline Jupiter or some deity of later introduction and non-Roman origin. It is also borne out by another significant and interesting fact--that the next image to be introduced, that of Diana in the temple on the Aventine, was a copy of the [Greek: xoanon] of Artemis at Massilia, itself a copy of the famous one at Ephesus.[295] Let us note that these two earliest statues were placed in roofed temples which were the dwelling-places of gods in an entirely new sense; so far no Roman deity of the city had been so housed, because he could not be thought of in terms of human life, as visible in human form