The same objection holds good as against Polites’ first view[585]. Taking the word λυκοκάντζαρος as his starting-point, instead of the common and central form καλλικάντζαρος, he proposed to derive the word from λύκος, ‘wolf,’ and κάνθαρος, ‘beetle.’ But though the resulting hybrid might be a monster as hideous as the worst of Callicantzari, these creatures so far as I know show no traits suggestive of entomological parentage. But since Polites himself has long abandoned this view, there is no need to criticize it further. His next pronouncement on the subject[586] banished both wolf and beetle and seemed to recognise the necessity of keeping the main form καλλικάντζαρος to the fore. But while he naturally assumed καλός to be the first half of the compound, he could only set down κάντζαρος as an unknown foreign, perhaps Slavonic, word. But in his latest publication[587] he relinquishes this position and falls back once more on a dialectic form καλιτσάγγαρος which is reported to be in use at the village of Pyrgos in Tenos and at some places on the western shores of the Black Sea. This word he believes to be a compound, of which the second half is connected with a Byzantine word τσαγγίον, meaning a kind of boot, and the still existing, if somewhat rare, word, τσαγγάρης, ‘a boot-maker,’ while the first half is to be either καλός, ‘fine,’ or καλίκι, ‘a hoof[588].’ The former alternative provides easily the form καλοτσάγγαρος or, as would be almost more likely, καλλιτσάγγαρος, meaning ‘one who wears fine boots’; while in the other alternative there results a supposed original form καλικοτσάγγαρος, meaning ‘one who has hoofs instead of boots,’ whence, by suppression of the third syllable, comes the existing word καλιτσάγγαρος, or again, by loss of the first syllable, a supposed form λικοτσάγγαρος which developed into λυκοκάντζαρος. On the score of formation the former alternative is unassailable; but the latter, with its supposed loss of syllables, is more questionable. The loss of a first syllable is common enough in modern Greek, where it consists of a vowel only (e.g. βρίσκω[589] for εὑρίσκω, μέρα for ἡμέρα, etc.), but the supposed loss of the syllable κα would, I think, be hard to parallel. Again the loss of a syllable in the middle of a word is fairly common either through the suppression of the vowel ι (or η, which is not distinguished from ι in sound) as in καλκάντζαρος for καλλικάντζαρος, ἔρμος for ἔρημος, etc., or else when two concurrent syllables begin with the same consonant, as in ἀστροπελέκι, ‘a thunderbolt,’ for ἀστραποπελέκι, but the loss of the syllable κο from the form καλικοτσάγγαρος is a bold hypothesis.