[The so-called prothetic vowel in Greek finds its ultimate explanation in Indo-Hittite. In Indo-Hittite bases beginning with the first, third, or fourth laryngeal, the first syllable was retained everywhere if it was in the full grade. If it was in the reduced grade, it was lost everywhere except in Hittite, Greek, and Armenian. If the second consonant was w, the first three laryngeals could unite with it to form a long voiceless hw which appears as spiritus asper in Greek; this accounts for the Attic and Homeric alteration of δvov and δvov. If the initial syllable began with the second laryngeal (;) and was in the reduced grade, it was lost everywhere, but the laryngeal united with a following liquid or nasal to give the long voiceless λ, , which ‘make position’ in Homer.]

The oldest HG documents that are preserved to us show a variety of unstressed final vowels: -i, -e, -a, -o, -u, and the diphthong -iu, e.g. nom. sg. hirti ‘Hirt’, dat. sg. tage ‘(dem) Tage’, nom. sg. zunga ‘Zunge’, haso ‘Hase’, fridu ‘Frieden’, guotiu ‘gute’ (fem.). At a later period most documents show instead of these full vowels only a single reduced vowel that is written -e: hirte, tage, zunge, hase, vride, guote (MG only; the UG form is guotiu). The same change can also be observed where the unstressed vowel is followed by a final consonant: early nom. acc. sg. neut. guotaz ‘gutes’ appears later as guotez, etc.

[IH ъA became IE ə. IH ъ? became Hitt. a; IH ъx became Hitt. aḫ(ḫ); IH ъγ became Hitt. aḫ before a vowel, a before s.]

It is a little over a hundred years (1837) since K. Zeuss first considered the possibility of a closer linguistic relationship between Germanic and Slavic. He gave the name ‘Slovenic’ to the language of the Slavs (or of the Wends, who according to him were ‘one people with the Slavs'), and maintained that the linguistic relationship which existed between that language and Germanic was no less close than that between Greek and Latin: ‘Not only have the two languages certain roots in common which are not found elsewhere; they also evince similarities in root adaptation, especially in flexion, as with none other.' From these facts he then draws the inference that ‘in the dim era into which not a ray of history penetrates’ the respective peoples, the bearers of the languages, 'regardless of whether their particular relationships of contiguity have been continued or not, have been in closer association with one another than others in whose case a more divergent tendency is noted in the unfolding of the foundation common to them.’ He likewise was of the opinion that 'at the other end . . . Slavic together with its Aistic or Baltic sister tongue’ is closely related to Indic.

[It is assumed in this paper that Indo-Hittite had four laryngeal consonants, namely: (1) a glottal stop of palatal color ('), (2) a glottal stop of velar color (;), (3) a voiceless velar spirant (x), and (4) a voiced velar spirant (γ), and that all of these consonants were lost in Primitive Indo-European. The paper attempts to prove that any one of the first three laryngeals converted an immediately preceding voiceless stop into the corresponding voiceless aspirate in Primitive Indo-European.]

[The paper illustrates the multiple relations of linguistics and ethnology by several discrete problems. See the last paragraph.]

[This paper outlines the uses, in Central Algonquian, of a peculiar morphological element, intermediate between root and suffix.]

[IE pure N-grades of the and types yield Alb. ul/lu and ur/ru, except where Ә appears to have been inserted (perhaps because the form in the N-grade was difficult to pronounce). The extended grades arising in this way fall together with original R-grades, which on becoming stressed or restressed yield Albanian combinations with the vowel a. IE N-grades of the and types are indistinguishable from extended grades and R-grades; they yield âm and ân in the Geg dialect, ëm and ën in Tosk. The article presents the evidence for these developments in a large number of etymologies, many of them new.]

[This article assumes that the Germanic phenomenon known as Verschärfung was due to the IE long voiceless semivowels hj and hw with an immediately following accent. These semivowels, in turn, resulted from an IH laryngeal consonant immediately preceding IH w or j.]

[An argument against Sturtevant's interpretation of the aspiration by laryngeals in the IH personal endings (Lang. 16.179–82); summarized in the last section.]

[The existence of visarga as a phonetic reality has been both accepted and denied. Evidence in the Prātiśākhyas points to the phonetic existence of several spirants written ḥ. These spirants, jihvāmūlīya and upadhmānīya, are shown to be allophones of visarga, which in turn is interpreted as a subclass of the s phoneme.]

[Quintilian has been thought to say that Greek φ was transliterated by Latin f. This is quite out of line with the findings of linguistic scholars concerning his period. An attempt is made here to prove that Quintilian's testimony is really not incompatible with these findings, but that Latinists have misunderstood him as a result of faulty emendation and interpretation.]

[Bloomfield has set up the Proto-Algonquian clusters *xk and *xp, using x to represent an undetermined phoneme preceding k and p. This paper proposes to show that *xk consists of two clusters in the parent language, namely *θk and *xk. In addition to establishing the validity of a new Proto-Algonquian cluster, it gives the reflexes for the principal Eastern Algonquian languages and postulates the stages of development.]

[The paper follows the development of the suffix -ivu in classical and medieval Latin and in the Romance languages, with the purpose of establishing the syntactic and semantic changes which it underwent and of determining the channels through which it filtered down from Latin into the vernaculars. Special attention is devoted to the substantival use of -ivu, as in Fr. l'objectif, l'initiative. See the summary at the end of the article.]