I am not sure why the -s is sometimes missing in original masculines (such as the "kettle" word). Younger loans, perhaps, or just the usual messiness in loanword endings. Its absence is to be expected in original Germanic neuters (like <patja> from *bad-ja-n, OE bedd, OHG betti, Gothic badi [n.], as opposed to ON beðr [m.]).
The "mother" word <äiti> has Gothic and West Germanic cognates (OHG eidî), and this fact seems to guarantee its PGmc origin, but I don't think it is found in the documented Scandinavian languages. It would seem that it was borrowed into Proto-Baltic-Finnic from PGmc or perhaps Old Runic (which I'd identify more generally with Proto-NW-Germanic rather than exclusively Proto-Norse). Some loans in Finnish are clearly older than attested Old Runic; for example, <rengas> shows no raising of *e to *i before *n, and that's a very archaic state of affairs even within Proto-Germanic. This implies that some of the loans must date back to well before 200 AD.
What we find in <ruhtinas> is not preaspiration but a separate segment corresponding to the PGmc. *x (velar fricative) in *druxt-i:n-a-z. You can see it in Old English dryhten 'lord' (same word, of course), where the articulation was still velar. Again, OE and Finnish reflect the original cluster slightly more faithfully than Icelandic does.
Finnish in general preserves the vocalism of early loans very well, but simplifies initial clusters (banned in Finnish phonotactics), so e.g. *fl-, *dr-, *xr- > l-, r-, r-, and adjusts the voicing and/or tenseness of Germanic obstruents in accordance with its own phonotactic demands. This is why *badja- became <patja>.
The number of old Germanic loans in Finnish is so high, and their semantics so characteristic (government, social institutions, technical terms, housekeeping, but also an occasional kinship term) that they strongly suggest something like a Germanic-speaking élite among the Baltic Finnic peoples about 2000 years ago and perhaps even earlier than that, not unlike the Normans in England. There is an interesting discussion of this in Raimo Anttila's (1972) _Introduction to Historical and Comparative Linguistics_ (New York: Macmillan), and in less readily accessible Finnish sources.
Hope we aren't moving too far OT.
----- Original Message -----
From: Óskar Guðlaugsson
To: norse_course@yahoogroups.com
Sent: Thursday, March 15, 2001 1:31 PM
Subject: Re: Odp: [norse_course] Finnish loans (was: synonyms)

--- In norse_course@......, "Piotr Gasiorowski" <gpiotr@......> wrote:
> A self-correction: Finn. kuningas, like <konung>, etc.,
means 'king', not 'prince'. The retention of the thematic vowel, the
old (unrhotacised!) Nom.sg. ending, and absence of any kind umlaut
make <kuningas> look very old. Here are a few other characteristic
loans (the list is far from complete):
> äiti 'mother' < *aiti:(n)- (Gothic aiþei)
> kattila 'kettle' (no umlaut)
> patja 'mattress' < *badja- (no umlaut, -j- preserved)
> ruhtinas 'prince' < *druxtinaz (all vowels preserved)
> rikas 'rich'

beðr (?)

Why is it not "kattilas"? Younger loan?

I suppose Finnish has no [b], explaining why *badja gets
rendered "patja".

The *aiþi:(n) word I cannot figure out; what is its descendant in ON?

Is "ruhtinas" actually pronounced [ruhtinas]? That is, is there pre-
aspiration there?

Last, I think some here would like some more explanations of what we
are talking about; the source forms (the ones from which the Finnish
forms are thought to derive) are reconstructed forms (often from the
Finnish forms themselves) of "Proto-Norse", the language preceding
ON. If I remember right, we're looking at the period 200-500 AD (?).
I trust Piotr will correct my dating. In any case, this is also the
language of the earliest rune inscriptions, right?