"mun-gat" manGad = "ant" aunt:  Grey 1840 [:202:32] [NYUNGAR]
"Mân-gat" manGad = "Aunt" aunt: Symmons, Charles [:4:4] [NYUNGAR]
"Man-gat" manGad = "Aunt; mother-in-law." aunt: Moore 1842 [:69:3] [NYUNGAR]
"man-gat" manGad = "mother-in-law" mother-in-law:  Symmons 1841 [:327:35] [Wajuk]
In the database from which the above extracts are taken there are ten examples similar to the above entries. All of the ten are provided by Grey, Moore or Symmons, and of these, Grey (1840), is the earliest. As in much of the database, it seems that one source copied copiously from another.
As can be seen from the first record above, Grey stated that 'manGad' meant 'ant'. Subsequently all family relationship interpretations of 'mangad' (for it had other quite different meanings too) were not 'ant' but 'aunt' or 'mother-in-law', i.e. senior female relatives. On the basis of 9 to 1, I opted for 'aunt' for Grey's word, assuming he had made a recording error. But perhaps he was right, and the copyists were wrong in their transcriptions of his work. When once an error is made, further copying compounds a blunder.
Sydney is far away, but nevertheless there is at times some relationship among the indigenous languages. All the Pama-Nyungan languages of most of the continent had a common origin. Over immense time, as the peoples spread over the continent, separate languages evolved.
In the Sydney language (which I have termed Biyal Biyal, abbreviated to BB), one word for 'ant' is 'mang'—close to 'mangad'.
"Mong" mang = "Ant ..." ant: Painters [:] [BB]
"Mong" mang = "Small brown ant" ant: Brown, Rbt: Georges R [264.72:2] [DgR]
"Mon" man = "Any" ant: Anon (c) [c:31:15] [BB]
"Mong" mang = "An ant" ant: King in Hunter [409.2:19] [BB]
"muun" mun = "ant (green)" ant: KAOL Ridley [KML] [20:18] [Kamilaroi]
"mu-un" mun = "Greenheaded ant" ant: Mathews KML/Dwl [278.5:9] [Kamilaroi]
"Moon" mun = "Small black ant" ant: SofM 1900 05 21 [Tibbetts] [63:38] [WIRA]
Similar words are found in the inland NSW languages Wiradhuri and Kamilaroi, as can be seen from the bottom three examples above.
Of course this might just be a coincidence: 'mangad' in Nyungar, 'mang' or 'mun/man' in certain NSW languages.
And finally there is another curiosity concerning 'ant'. In the 'Anon' notebook record of BB vocabulary compiled around 1790 a transcription error was made in copying a rough note into the governor's word list, as the middle item in the following extract shows: