Participles -- Two Questions

(A) τιθείς

Yes, you have it right (except for the accent).

Athematic (-μι) verbs have both long and short stems (e.g. the long grade τιθη- vs. the short grade τιθε-). The ptcp uses the short stem. In fact, any verb-stem ending in a long vowel will usually shorten in the active ptcp. This is because the active ptcp is formed with the suffix ντ, a consonant cluster beginning with a resonant (μ ν ρ λ). Long vowels shorten before resonant-initial consonant clusters (this is called Osthoff’s Law). E.g. the aor pass ptcp πεισθέντες<*πεισθήντες (η shortens to ε before ντ).

(B) τιθεῖσα

You’re actually working with ντ and the semivowel yod (y) [see this thread for a little more detail on yod: http://discourse.textkit.com/t/topic/17490/8]

The feminine ptcp is formed with the suffix -yα (short alpha): *τιθεντyα. Tau + yod sometimes results in sigma, so *τιθεντyα>*τιθενσα. The nu drops out with compensatory lengthening: *τιθεντyα>*τιθενσα>τιθεῖσα.

Cp. πᾶσα<*πανσα<*παντyα