Excerpt from change request document:
In my 2004 Dogon survey report I wrote: "According to the elders of the Mombo speaking village Bandiougou, Ampari-kora speakers could understand them, but not vice-versa. One explanation might be inherent intelligibility, since Calame-Griaule listed both Mombo and Ampari-kora (she called it Ãmbaŋge) as subvarieties of Edyeŋge dõ. However, since the Ampari-kora speakers that the elders spoke about came yearly to Bandiougou to observe a rite, this might also be learned comprehension."
In 2009 I analysed the following Dogon wordlists in Wordsurv 6.0.2: Miambo de Makou, Mombo de Bandiougou, Mombo Kɔlu Sɔ, Mombo de Kori-Kori, Mombo Pignari, Nyambaleeŋge de Ficko, Mombo (Blench's Ambaleenge) de Kema, and Ampari de Pahandaga, Ampari de Pa, Ampari de Danga, Ampari de Baboye. Raw scores between Mombo varieties range from 71-88%. Raw scores between Ampari varieties range from 70- 88%. Raw scores between Mombo and Ampari varieties range from 40-53%. Total number of items compared range from 64 to 180.
"...Mombo and Ampari are separate non-intercomprehensible varieties, which, however, are genetically much closer to each other than to any other Dogon language." (correspondence with Prokhorov 12-19-2009)