cf. Wikipedia:WikiProject Languages/Glottolog 3.3 language names

Hammarström (2015) Ethnologue 16/17/18th editions: a comprehensive review: online appendices

Missing languages

Africa

Data in Klieman (2003, 1997), Sato (1992) of the language spoken by the Mikaya, Bambengangale and Baluma in the Republic of Congo are so different from other languages in the Likouala-Sangha Bantu subgroup that a separate entry is required. These lects are not covered in any form in any other entry in Eth18.
According to van der Burgt (1902), the language of the Kitwa among the Rundi speak a language of their own (in addition to Kirundi itself) which is not understood by the neighbouring Watussi and the Wahutu speakers and has the same grammar as Kirundi but is lexically different (the bulk of which is nevertheless of Rundi origin!). This language is missing. per later sources this is Rundi
Belueli, a Pygmy language spoken at Apare in the Ituri forest area of NE Congo, is missing. It is not intelligible to neighbouring Bali [bcp] and appears to the in the Nyali-Ndaka-Mbo-Budu-Vanuma subgroup (Schebesta 1953).
The Gabonese Bantu language Mwesa (Mouloungui 1999) is missing. It is as distinct from all other Gabonese languages as any other Gabonese language included in Eth18 (Alewijnse et al. 2007, Idiata 2007).
The Gabonese Bantu language Tombidi (Bingoumou 2005, 2008) is missing. It is as distinct from all other Gabonese languages as any other Gabonese language included in Eth18 (Alewijnse et al. 2007, Idiata 2007).
The Ngala language of Lake Chad, as documented in a wordlist by Barth (Benton 1912, von Duisburg 1914), is missing. It is clearly of the Kotoko group of Chadic, closest to Makari (= Mpade [mpi]) and Goulfey (= Malgbe [mxf]) but is as different from both as Makari and Goulfey are from each other (Migeod 1922, Sölken 1958).
A survey of Fadan Karshe-Wamba-Akwanga triangle in Kaduna State, Nigeria revealed the existence of four extinct or near-extinct unknown languages Akpondu, Babur, Nisam and Nigbo (Blench 2005a), none of which are in E16. A little data exists to confirm that Akpondu was indeed a different language from all others (Alumu-Tesu [aab] being the closest relative), but for the others there is so far only sociolinguistic information which not sufficient to totally exclude that Babur, Nisam and Nigbo were unintelligible to themselves and all other languages. I wish to thank Roger Blench for drawing my attention to Akpondu.
The language of the villages Kazibati and Mongoba (likely belonging to the Ngbandi group, see van Bulck 1948:178, Hackett & van Bulck 1956:74, McMaster 1988:123, 259), is missing (Costermans 1938).
Ham, an endangered language of the North Masa-Musey group (Melis 2006:45), is missing (Ajello et al. 2001). Ham is not intelligible to Masa [mcn] though many Ham speak Masa (Melis 2006:44).
Gizey, a language of the North Masa-Masa group, is missing (Ajello 2006, Ajello et al. 2001). Gizey is not intelligible to Masa [mcn] though many Gizey speak Masa (Melis 2006:45).
Julud is mutually intelligible with Katla-Kulharong but defintely not with Katla-Cakom, so it should be listed as a separate language from Katla [kcr] (p.c. Birgit Hellwig 2007, N/A 1979a,b)
Lwel or Kelwer, a Bantu B862 language of the Democratic Republic of Congo is missing (Khang Levy 1979). It is not mutually intelligible with any of its neighbours (Koni Muluwa & Bostoen 2011).
Mangar, a Chadic language of the Ron group, is clearly different enough from Da�o-Butura (and other Ron languages) to be a separate language (Blench 2001) and is duly listed as a separate language in informed reference works (Jungraithmayr 1981).
Shang, a Hyamic language of Nigeria, is missing (Blench 2010b).
Dyarim, a Chadic language of the West Chadic Zaar group, is missing (Blench 2007).
The Bafotó language of Pygmoids among the Mongo (not intelligible with Mongo) is missing (Hulstaert 1960, 1978) and is highly endangered if not extinct (Motingea Mangulu 2001-2002).
The now probably extinct (Cobbinah 2013:32) Northeast Bainounk varieties of northeast Casamance (once spoken in the villages Kamanka, Kansambu, Diouroul, Velingara and Samakoung) are missing in Eth18 either as part of other entries or as an entry of their own. Since they were only about 60% lexicostatistically cognate (Doneux 1991:87) to the other variety of Bainounk, they must be inferred not to have been intelligible to other Bainounk speakers5.
Syan (glossonym OruSyan, Huntingford 1965) is a missing language of the North Nyanza subgroup of Bantu, which lexicostatistically too divergent to be intelligible any other language in Eth18 (Schoenbrun 1994). merged per Maho

SAm

Wayumara, an extinct Guianan Cariban language most closely related to Maquiritare [mch], is missing (Farabee 1924, Girard 1971, Koch-Grünberg 1928b, Schomburgk 1849, von Martius 1867b).

Eurasia

Older Runic (Antonsen 1975) --cross-ref'd from Proto-Norse

Spurious languages

Eurasia.

H couldn't attest that Pu Ko [puk] actually exists. Should the article be retired? Thanks, — kwami (talk) 00:09, 23 March 2016 (UTC)

Also Southern Lolopo [ysp], and Kuanhua [xnh] which I assume is legit because you've edited the article. Thanks. — kwami (talk) 00:31, 23 March 2016 (UTC)

I have no idea what Kuanhua is to be honest, and I have not found any autonyms or non-Chinese names that could correspond to it. As for Pu Ko, I would vote to keep it there. — Stevey7788 (talk) 13:39, 23 March 2016 (UTC)

Okay, Southern Lolopo is merged with Lolopho.

Do you have a page number where Li (2005) attests to Kuanhua?

The pages right before the word list appendix. I'll see if I can look it up. — Stevey7788 (talk) 03:59, 30 March 2016 (UTC)

Is Lopi [lov] distinct or even attested?

Some sort of Northern Yi lect. I'd say keep it. — Stevey7788 (talk) 03:59, 30 March 2016 (UTC)

Is Lama [lav] a variety of Nung [nun], or is it the Laemae (lɛ˨˩mɛ˨˩) within the Bai branch? (It currently redirects to Bai.)

It's a mixed language, cf. Bradley's papers. — Stevey7788 (talk) 03:59, 30 March 2016 (UTC)

Maybe outside your area, but can you confirm if Inpui/Kabui Naga [nkf] and Puimei Naga [npu] are varieties of Rongmei Naga [nbu]?

Not exactly outside my area, I've been doing some work on NE Indian languages as well. I believe those should all get their own articles. However, they're all Zeme languages. — Stevey7788 (talk) 03:59, 30 March 2016 (UTC)

E18 Norra language [nrr], is listed as a Nungish language of Northern Myanmar (no more precise location) with the dialects Nora, Byabe, Kizolo. The information almost certainly derives from Voegelin & Voegelin (1965:15), but Voegelin & Voegelin (1965:15) list Nora, Byabe, Kizolo as dialects of Nung language (which has an entry Nung [nun]). Indeed, eldwork conrms Nung in China and Nung across the border in Myanmar are mutually intelligible (Sun & Liu 2009:8). Khamyang per sources
E18 Pu Ko language [puk] is listed as a SW Tai language in Laos without a speaker estimate or a location. No such language or ethnicity could be found in the comprehensive ethnic survey of Laos (Schliesinger 2003c:287).
E18 has an entry Shuadit language [sdt] for an extinct allegedly Jewish variety of Provençal. But Provençal spoken by Jews has been investigated and found to be indistinguishable from non-Jewish varieties of Provençal language (Banitt 1963, Pansier 1925, Guttel & Aslanov 2006:560) which already have entries in E18.
E18 has a Mon-Khmer, Unclassied entry Kuanhua language [xnh] with alternative name Damai, located in Southwest Yunnan Province, Xishuangbanna Dai Autonomous Prefecture, Jinghong County of China, with the comment that they are locally considered part of Khmu [kjg]. This entry does not seem to match any language known in the literature, although there are many languages in Yunnan and the entry has too little information to decide this properly. One possibility is that Damai is rendering of tmOOi 'subgroup' and that Kuan is a rendering of kwEEn (an ethnonym) that passed through a Chinese transcription. If the Khmu kwEEn on the Chinese side of the border were asked are who they are, a plausible response would be tmOOi kwEEn 'the kwEEn subgroup'. I owe this ingenious suggestion to Jan-Olof Svantesson (who himself calls it a 'wild guess', p.c. 2011). If the guess is correct, the Kuanhua [xnh] entry is just Khmu [kjg] spoken on the Chinese side of the border. If not, Kuanhua [xnh] entry should be removed on the grounds of insucient information to decide if it is the same as an otherwise existing entry.
E18 has an entry Lopi language [lov] as an unclassied Loloish/Ngwi language of Myanmar (E16)/China (E18). The only unique reference to such an ethnic group is the mention by Madrolle (1908:544) (as Lo-pi, No-pi or Nou-pi). It is perfectly possible that Madrolle (1908:544) means that the Lo-pi speak a variety of the language in E18's Honi [how] entry, and there is nothing to assert that the Lopi speak a language mutually unintelligible to Honi [how] and other Hanic languages.
E18 Lama language (Myanmar) [lay] is a name for a section of Nung language [nun] of Nungish branch of Tibeto-Burman (Voegelin & Voegelin 1965:15).
E18 The Inpui Naga language [nkf] entry, also known as Kabui language, is another name for a variety in the Naga-Rongmei language [nbu] entry (Marrison 1967:II:361-362, Grierson 1903).
E18 The entry Puimei Naga language [npu], as described in E16, refers to the same language as the Naga-Rongmei language [nbu] entry (Kabui 2004). There may be dierences within the Naga-Rongmei [nbu] entry that justify dividing it, but the division should, in that case, not be Puimei vs Inpui vs Naga-Rongmei.