This root and ones like it were used for Elvish words for “silver” throughout Tolkien’s life. The earliest iteration of the root began with T-, however, appearing as unglossed ᴱ√TELEPE in the Qenya Lexicon of the 1910s with derivatives like ᴱQ. telpe “silver” (QL/91). Even at this early stage, however, the Gnomish equivalent was G. celeb (GL/25), but the reason for the t/c variation isn’t clear. The closest explanation is that palatal consonants like [c] became [tʲ] in Qenya vs. [k] in Gnomish (compare ᴱQ. tyava- vs. caf- “taste” from ᴱ√TYAVA) but this doesn’t explain why the Qenya form has initial t- rather than ty-.
Elsewhere in the Elvish languages of the 1910s there seem to be etymological variations of [k] vs. [t], such as ᴱQ. kitya- vs. G. tisca- “tickle” (QL/47; GL/70) and ᴱQ. talqe vs. G. celc “glass” (QL/88; GL/25), so perhaps ᴱQ. telpe vs. G. celeb “silver” is another example of this. Another explanation appeared in Early Noldorin word lists from the 1920s, where the primitive form was ᴱ✶kelekwé which produced ᴱN. celeb as usual but the Qenya form was ᴱQ. telqe with “k = t by dissimilation” (PE13/140), presumably away from q.
In The Etymologies of the 1930s Tolkien had the root ᴹ√KYELEP with variant ᴹ√TELEP, producing N. celeb but ᴹQ. tyelpe or ᴹQ. telpe (Ety/KYELEP). But Tolkien revised this entry, marking ᴹ√TELEP as questionable and introducing the Telerin form ᴹT. telpe < ᴹ√KYELEP, concluding that ᴹQ. telpe must be a loan from Telerin. This finally put N. celeb vs. ᴹQ. telpe (borrowed from Telerin) on a solid phonological foundation. Tolkien seems to have stuck with this explanation, mentioning this borrowing from Telerin to Quenya several times in his later writings, with the proper but now archaic Quenya form being Q. †tyelpë (Let/426; PM/356; UT/266).
Tolkien used words beginning with tel- or tyel- for “end” and these were intermixed with “roof” words, but the exact arrangements evolved over Tolkien’s life. In the Qenya Lexicon of the 1910s, the base root was ᴱ√TELE “cover in” with derivatives like ᴱQ. tel (teld-) “roof” and ᴱQ. telimbo “canopy; sky”, along with a vocalic extension ᴱ√TEL+U “to finish, close, end, complete” having derivatives like the verb ᴱQ. telu- of the same meaning and ᴱQ. telwa “last, late” (QL/90-91). Both the base root and its extension had derivatives in the contemporaneous Gnomish Lexicon such as G. teld “roof” vs. G. telu “end” (GL/70).
In the Early Qenya Word Lists of the 1920s, however, Tolkien gave the words ᴱQ. talle “ended” and ᴱQ. talma “end” (PE16/143, 144) versus ᴱQ. telu- “cover, roof” and ᴱQ. telume “firmament” and (QL/134, 142). Indeed, in The Etymologies of the 1930s Tolkien had unglossed ᴹ√TEL with the vocalic extension ᴹ√TELU having derivatives like ᴹQ. telme “hood, covering” and ᴹQ. telume/N. telu “roof (of heaven)” (Ety/TEL). The primitive verb form ✶telu- “roof in” appeared in Quendian & Common Eldarin Verbal Structure (EVS1: PE22/98) from the 1930s and again in Common Eldarin: Verb Structure from the early 1950s (EVS2: PE22/135), and finally one more time with the gloss “roof in, put the crown on a building” in the Quendi and Eldar essay of 1959-60 (WJ/411).
The “end” words went down a different route. In The Etymologies Tolkien gave ᴹ√KYEL “run out, come to an end” as an etymological variant of ᴹ√KEL “go, run (especially of water), ✱flow”; ᴹ√KYEL had derivatives like ᴹQ. tyel- “to end, cease” and ᴹQ. tyelima “final” (Ety/KEL, KYEL). Tolkien mentioned this etymological variation of √KEL “flow” vs. √KYEL “cease, come to an end” in both the first and second version of Tengwesta Qenderinwa from the 1930s and around 1950 respectively (TQ1: PE18/58; TQ2: PE18/103), and the two variants were also mentioned in the Quenya Verbal System (QVS) from 1948 (PE22/114). However, in Common Eldarin: Verb Structure of the early 1950s Tolkien gave √TEL “close, end, complete” with vocalic extension telu “roof in” (PE22/135), a paradigm repeated in the Quendi and Eldar essay of 1959-60, though with a bit less certainty:
> The stem ✱TELE, the primary sense of which appears to have been “close, end, come at the end” ... This was possibly distinct from ✱tel-u “roof in, put the crown on a building”, seen in Q telume “roof, canopy” ... But ✱telu may be simply a differentiated form of ✱TELE, since the roof was the final work of a building (WJ/411).
The conceptual development seems to be 1910s ᴱ√TELE “cover” vs. ᴱ√TEL-U “end” >> 1920s ✱ᴱ√TALA “end” vs. ᴱ√TELU “cover” >> 1930s-1950 ᴹ√KYEL “end” vs. ᴹ√TEL(U) “✱cover” >> early 1950s √TEL “end” vs. √TELU “roof in”. Thus Tolkien ultimately came full circle back to his original root forms, but with reversed meanings.
See also the root √TELES for discussion of the conceptual shifts in the derivation of Q. Teleri “Hindmost”.