An initial s followed by a voiced stop developed into a voiceless spirant of the same quality: sp, st, sk > f (or ꝑ), þ, χ (IPA [f] or [ɸ], [θ], [x]). A similar change occurred in Sindarin. Tolkien discussed this sound change in both the Outline of Phonetic Development (OP1) from the 1930s and the Outline of Phonology (OP2) from the 1950s:

[Initial] sky, skw became simplified to [χy, χw] whence Q hy, hw, identical with the products of khy, khw: see above. The same development was common in the case of sp, st, sk which thus became [ꝑ, þ, x] > f, th (s), h initially (OP1: PE19/30).

[Initial] sp, st, sk. For these initial groups a development to f, þ, h identical with that of ph, th, kh but probably coalescing with these only at the spirantal stage, is well evidenced, & occurs in many words and stems. It is improbable that sp, st, sk ever became aspirates. They more likely reached the stage f, þ, h through long spirants ff, þþ, χχ soon simplified (OP2: PE19/78).

In the 1930s, Tolkien described similar developments for initial sky, skw, but in the 1950s he decided these combinations did not occur initially in Primitive Elvish (PE19/78 note #51). The Quenya phonetic developments were a bit different than in Sindarin/Noldorin, where it seems that st- > sþ- > þ-, but in Quenya st- > þþ- > þ-. As indicated by the quotes above, it seems this sound change was after aspirates became voiceless spirants, but Tolkien seems to have vacillated on whether it was before or after [ɸ], [β] became [f], [v], so whether sp- > ꝑ or f is unclear.

Attested examples are quite numerous in the 1930s, 50s and 60s, and some a few examples appear as early as the 1920s, such as: ᴱ✶skantá [> *χanta] > ᴱQ. hanta “blow with an axe” (PE13/147) and ᴱ✶stak+ta [> *þakta] > ᴱQ. sahta- “to split” (PE14/66). It seems plausible the same sound changes applied to the earliest Qenya of the 1910s but there are no clear examples.