This is a very good example of the philological way of explaining a myth. If once we admit that _ark_, or _arch_, in the sense of 'bright' and of 'bear,' existed, not only in Sanskrit, but in the undivided Aryan tongue, and that the name Riksha, bear, 'became in that sense most popular in Greek and Latin,' this theory seems more than plausible. But the explanation does not look so well if we examine, not only the Aryan, but all the known myths and names of the Bear and the other stars. Professor Sayce, a distinguished philologist, says we may not compare non-Aryan with Aryan myths. We have ventured to do so, however, in this paper, and have shown that the most widely severed races give the stars animal names, of which the _Bear_ is one example. Now, if the philologists wish to persuade us that it was decaying and half-forgotten language which caused men to give the names of animals to the stars, they must prove their case on an immense collection of instances--on Iowa, Kaneka, Murri, Maori, Brazilian, Peruvian, Mexican, Egyptian, Eskimo, instances. It would be the most amazing coincidence in the world if forgetfulness of the meaning of their own speech compelled tribes of every tongue and race to recognise men and beasts, cranes, cockatoos, serpents, monkeys, bears, and so forth in the heavens. How came the misunderstood words always to be misunderstood in the same way? Does the philological explanation account for the enormous majority of the phenomena? If it fails, we may at least doubt whether it solves the one isolated case of the Great Bear among the Greeks and Romans. It must be observed that the philological explanation of Mr. Müller does not clear up the Arcadian story of their own descent from a she-bear who is now a star. Yet similar stories of the descent of tribes from animals are so widespread that it would be difficult to name the race or the quarter of the globe where they are not found. Are they all derived from misunderstood words meaning 'bright'? These considerations appear to be a strong argument for comparing not only Aryan, but all attainable myths. We shall often find, if we take a wide view, that the philological explanation which seemed plausible in a single case is hopelessly narrow when applied to a large collection of parallel cases in language of various families.