Since Usener put prominently forward the theory of Sondergötter, the idea has been subjected to trenchant negative criticism by two experts in the fields of Greek and Roman religion respectively, Farnell and Wissowa. The former protests, and rightly so, against the cheerful assumption that, whenever a deity has a name describing a function, “Saviour,” “Queen,” “Victory,” for example, we should regard him or her as a primitive Sondergott; since many examples teach us that such figures are often the products of a developed polytheism. He would suggest, as a better test than the name, the non-anthropomorphic conception of the god, or rather daimon, in the minds of his worshippers. Thus he clears the field of Greek religion of a great many heroes and daimones who, whatever their names may be, are too developed and too late to have any claim to represent primitive thought. Wissowa attacks the question from a somewhat different standpoint. He sees in the formidable list of Roman Sondergötter nothing more recondite than Varro's attempt to arrange all possible deities under “di certi,” or at most the artificial “indigitamenta” of the pontifices which, in accordance with “die peinliche Genauigkeit in der Aufstellung der römischen Gebetsformeln,” endeavoured to call upon whatever god was addressed under all the names which applied to the actual petition.