I have tried to show (Language 3. 109–22) that Hittite h occurs where the Indo-European languages have bh initial, and I can now make two or three additions to the material there presented. In Language 4. 122 f. I have connected halanta ‘head’ with Greek φαλός·λευκός, φαλακρός ‘bald’, and Sanskrit bhālam ‘forehead’, on the assumption that the original meaning of the Hittite word was ‘bald’. Hittite haš-, hašh- ‘open’, must be identified with IE *bhosos ‘naked’, whence Lithuanian bãsas ‘bare-foot’, Old English bær, etc. The verb pihiš-, which, accompanied by the adverb arha, means ‘strike off, cut off, take off’ or the like, contains the verbal prefix pe-, while hiš- is an extension in s of the root which appears in IE *bhei- ‘strike’ (Old High German bīhal ‘axe’, etc.), and whose extended form *bheid- means ‘split’ (Sanskrit bhinadmi, Latin findo, etc). Another instance of h = bh after the verbal prefix pe- is pehar(k)- ‘hold towards’ (Keilschrifttexte aus Boghazköi 4. 2. 2. 25, Keilschrift-Urkunden aus Boghazköi 13. 4. 4. 37) beside har(k)- ‘have’ on which see Language 3. 117 f.

Hindus of reflective tendencies have always been much given to linguistic analysis and speculation. All Indian systems deal more or less with problems of the nature of language, the relation of sound to sense, etc. When the time comes for a general History of Linguistic Theories, the Indian section will bulk large. While none of the recognized philosophic systems can be neglected, the three most important ones from this standpoint are, no doubt, the Vaiyākaraṇikas or grammatical school (most prominently represented by Pāṇini), the Naiyāyikas or followers of Nyāya, and the Mīmāṅsakas or school of the Pūrva-Mīmāṅsā.

Certain forms of the Greek word for ‘mix’ which have γ instead of κ as the final of their root-syllable, e.g., the present μíγνυι and the 2nd aorist ἐμíγην, appear to be anomalous when compared not only with the other forms of the same Greek verb but also with its cognates in other Indo-European languages, where all forms point to as the primitive IE root-final. Boisacq derives μíγνυμι from pIE *mei- and states, without further comment, that ‘le grec a la gutturale douce’. Walde, s. v. misceo, gives the Greek cognate as ‘mit Media’, but makes no remark on the phenomenon except to reject Wackernagel's theory of an independent origin for μíγνυμι. Apparently, then, Boisacq and Walde explain the disagreement in root-finals by the theory of the socalled variation of the root-determinative.

Apparently without guiding principle, the perfect participles passive of certain verbs in Latin have a short radical vowel, while others have a long vowel: thus factus, spectus, scissus, to faciō, speciō, scindō; but āctus, rēctus, vīsus, to agō, regō, videō. An excellent summary of the discussion which has grown up around this point was given by Sommer in 1914; but for the proper understanding of the problem it will be necessary briefly to repeat the main points.

The object of this article is not to add one more hypothesis to the many others that have at various times been advanced to explain the origin of the Latin vī-Perfect, but rather to call attention to an old and neglected explanation which seems to the writer so simple and obviously superior to all the vague guesses which are apparently enjoying much greater favor, that it is a source of amazement, e.g. that Stolz-Leumann, Lat. Gram. 335, do not even mention the same in their extensive enumeration of theories and authorities.

In Language 4. 28–30, Prof. U. T. Holmes tries to show that some native Celtic word must have had some part in the etymology of caroler ‘to dance a round dance’. He refers to a Welsh coroli ‘to move in a circle’ and to a Welsh corol (or corawl) ‘circular’. But I doubt whether sufficient authority can be found for either of these words.