>Phrygian has <anar> 'husband', and Macedonian has <abroutes> 'eyebrows' (or <abrouwes> if you join Kretschmer in taking the tau as a copyist's error for digamma in Hesychius; <ou> is of course Late Greek for /u:/). On the other hand Aristotle's <sausarismos> 'hoarseness' shows that Mac. lost the initial laryngeal of *h2souso- 'dry, parched'.
> > The specific effects of the laryngeals are various, and show up
> > slightly differently in different languages. They include vowel
> > length and aspiration, but also accentuation, and various kinds of
> > paradigm irregularity.
> > Peter
> One thing I've been wondering about the laryngeal theory is whether laryngeal-induced coloring (and lengthening) is a comparativ or internal reconstruction. My current impression is that it's the latter - and so for PIE proper the situation would have been all six of *a *e *o *a: *e: *o:, tho we'll still need some actual laryngeals around to account for Anatolian. How Brugmann's Law fits here I'm not sure.
> Another note, is there any evidence outside of Greek for root-initial preconsonantal (ie. vocalizing) laryngeals? If not, is there anything speaking against considering these prefixes of some sort? The idea of various complex onset clusters reconstructed on the evidence of one branch only sounds distressingly ad hoc to me.