*d(i/u)l

This etymon is sometimes hard to distinguish from #300 PTB *d(w)əy EGG / TESTICLE, above. It is possible that conflation with an Indo-Aryan root is involved. Jäschke (p. 234) says that WT thul is “according to Cunningham a Cashmiri word”. The TB cognates are indeed confined to Indospheric branches of TB (NE Indian Areal Group and Himalayish).

Some Himalayish forms for HEEL have an element like -din- which look as if they could probably come from this etymon for EGG (see the note on Lushai (Mizo) ke-ar-tui): these include Kulung 'dhin-di-ri, and Thulung and Khaling din-di-ri. Yet the suspicious similarity among these forms suggest that they might be loans from Nepali.

This is very likely the same etymon as *r-tul ⪤ *r-til DULL / BUTTOCK / HEEL / ROUNDED PART (HPTB p. 419), cf. WT rtul-po ‘blunt, dull’; Abor-Miri ko-dun ‘buttock’; Meithei mə-thun ‘buttock’; Wancho chi-dun ‘heel’ (chi ‘foot’); Khözha šɯ́-dò; Lisu khi²¹du²¹ ‘buttock’ (khi²¹ ‘excrement’); Phunoi pi³³tun¹¹ ‘heel’. See Matisoff 1994 (“How ‘dull’ can you get?”), and the Chinese comparanda, below. The allofam with medial -i- is represented by Jingpho šətīn ‘buttock’, ləthīn ‘heel’, and also perhaps by WT rtiŋ-pa ‘heel’.

Benedict apparently had a different theory. He implies a connection between Lushai (Mizo) til ‘testicle’ and Thado til ‘earthworm; testicle’ (cf. also PLB *di ‘worm’ > WB ti, Lisu bi-di), which he reconstructs as PTB *zril ‘worm’ (STC, n.121, p. 37). Several Chinese comparanda meaning WORM are offered (n.457, p. 171). It must be said, however, that the semantic association between TESTICLE and WORM is a bit obscure.

There are several likely Chinese comparanda (HPTB:422, 504), including:

臀 OC *d’wən, GSR #429b-c ‘buttocks’
殿 OC *tiən, GSR #429d ‘rear of an army’
沌 OC *d'wən, GSR #427h ‘confused / stupid’
鈍 OC *d'wən, GSR #427i ‘dull’
頓 OC *twən, GSR #427j ‘worn / dull / spoiled’.
[JAM]

As noted above, if this PTB root is related to *r-tul ⪤ *r-til DULL / BUTTOCK / HEEL / ROUNDED PART, then it can be compared to the following Chinese word family (see Coblin 1986:67-68; Gong 1995 #154; HPTB:422, 504):

臀 OC *d’wən, GSR #429b-c ‘buttocks’; Li 1971: *dən; Baxter 1992: *dun; Mand. tún.
殿 OC *tiən, GSR #429d ‘rear of army’; Li 1971: *tiənh; Baxter 1992: *tɨns; Mand. diàn.
沌 OC *d’wən, GSR #427h ‘confused, stupid’; Li 1971: *dənx; Baxter 1992: *dunʔ; Mand. dùn.
鈍 OC *d’wən, GSR #427i ‘dull’; Li 1971: *dənh; Baxter 1992: *duns; Mand. dùn.
頓 OC *twən, GSR #427j ‘worn, dull, spoiled’; Li 1971: *tənh; Baxter 1992: *tuns; Mand. dùn.

The correspondence between PTB *-l and OC *-n is regular (see Gong 1995 for numerous examples). The initials and vowels also match well. The voicing alternation and suffixation seen in the Chinese word family are typical, although in this case the morphological function is not clear.
[ZJH]

The following comparanda are offered for PTB *zril EARTHWORM in STC p. 171:

蟺 OC *d̑i̯ɑn, GSR# 148p; Mand. shàn
蚓 OC *di̯ĕn, GSR# 371c; Mand. yǐn
螾 OC *di̯ən, GSR #450j; Mand. yǐn

[JAM]

蟺 OC *d̑i̯ɑn, GSR #148p ‘earthworm’; Li 1971: *djanx; Baxter 1992: *djanʔ; Mand. shàn.
蚓 OC *di̯ĕn, GSR #371c ‘earthworm’; Li 1971: *rinx; Baxter 1992: *ljinʔ; Mand. yǐn.
螾 OC *di̯ən, GSR #450j ‘earthworm’; Li 1971: *rənx ?; Baxter 1992: *ljɨ/inʔ; Mand. yǐn.

Benedict (STC p.37 note 121 and p. 171 note 457) argues that all three Chinese words are related and ‘point... to an original initial such as *zr-‘. These in turn are compared to TB *zril ‘worm’.¹ Based on this and a handful of other comparisons, Benedict argues for the following developments from PST to Chinese: *zr- > *źr- > *d̑i̯ ² varying with *zr- > *zy- > *y- > *di̯.

Based on our current understanding of Old Chinese, Benedict’s hypothesis is no longer sustainable, at least not in full. GSR #148 is a dental series while GSR #371 and #450 are lateral series. This and the vowel difference indicate that 蟺 shàn is not an allofam of 蚓 yǐn and 螾 yǐn.³ As for the latter two, the reconstruction in both Li’s and Baxter’s systems is difficult. The Middle Chinese forms might be descended from either OC *i or OC *ə (Li)/*ɨ (Baxter). Li and Baxter agree that GSR #371 has main vowel *i, but the reconstruction of GSR #450 is ambiguous.

Schuessler 2007:574 explicitly relates 蚓 yǐn and 螾 yǐn, indicating that they are variant graphs used to write the same morpheme. He reconstructs OC *lə/inʔ or *jə/inʔ. While various attested binomial forms for ‘earthworm’ suggest that *i is the OC vowel, Min dialect forms point to *ə.

As for the initial, it is now generally agreed that it should be *l or *j, not *r.

Looking again at Benedict’s comparison with PTB *zril, we note that the vowel and coda correspondences are regular. A comparison of PTB *zr- with OC *l- or *j- looks doubtful on phonetic grounds, but cannot be dismissed out of hand. Since *zr- is so rare in TB, it is difficult to establish regularity of correspondence.

Whatever the fate of that comparison, Benedict’s claim that these Chinese words are ultimately connected to #1181 PTB *d(i/u)l EGG / TESTICLE now seems quite unlikely to be true.
[ZJH]