Vatrata [vlr] is listed in E16 as a separate language, but this is simply a village name of the language covered in the Vera'a [vra] entry ( John Lynch and Terry Crowley 2001 :40 ). See also: Vera'a [vra].
Excerpt from change request document:
Vera'a and Lemerig belong historically to two distinct language communities, even though Lemerig speakers now have generally adopted Vera'a as their primary language. There is low intelligibility between Vera'a and Lemerig.
"Vatrata" is not, and never was, used by the local community. It was chosen by XIXth-c. missionaries based on another language (Mota). The speakers designate their village and language as Vera'a (apostrophe = glottal stop). This is also the spelling which I have adopted in my scientific work, and which has been used by other scholars ever since. "Vatrata" is used by nobody else than Codrington (1885). As for "Vetrat", it is the name of the place & language, in the dominant language of Vanua Lava (Vurës [msn]).
The dialect names "Leon, Pak, Sasar" as listed in the Ethnologue form part of the distinct language called Lemerig.
There is no body of literature either in or on either of these languages, no other reason to keep the 2 languages under a single code.
Vera'a should be considered the main language of the two, because it is viable and spoken by more speakers (about 300); Lemerig is dying out, with only 2 speakers.
Important note: the three dialects currently mentioned under "Vatrata", namely "Leon, Pak (Bek) and Sasar", correspond in fact to various ancient varieties of the language Lemerig. As far as I know, all these dialects became extinct during the last generation, except for Pak (or more precisely Päk) which still has 2 speakers alive. Everybody insist that these three names, all corresponding to small mountain hamlets now deserted, should be placed under the umbrella term "LEMERIG" to designate the language.
Finally, Vatrata has no internal dialect.