I found the statement which I called "Kosnita's Theorem" in a Rumanian
book; previously it had appeared in the Rumanian journal "Gazeta
Matematika". Perhaps Kosnita has been a Rumanian mathematician. However,
his name probably has a Bulgarian origin: it should be pronounced
"Koshnitza" and means "basket". Since Rumania and Bulgaria are
neighbouring countries, in the languages there are common words; may be,
there is such a word in Rumanian. The old Rumanian has been close to
Bulgarian, but in the second half of 19-th century it has been
"romanized" and has become rather different.
With best wishes, J. Tabov.
Julio Gonzalez Cabillon wrote:
> ...
> Do you happen to know anything about Kosnita [see below]? If so,
> I would much appreciate any information to share with interested
> listmembers. Could "Kosnita" be a place, or a transliterated
> Bulgarian word, instead of the name of a mathematician?
> ...
>
> ========
>
> In the September 1995 issue of the Mathematics and Informatics
> Quarterly, the following interesting result was mentioned without
> proof in the column on Forgotten Theorems:
>
> Kosnita's Theorem. The lines joining the vertices A, B, and C
> of a given triangle ABC with the circumcenters of the triangles
> BCO, CAO, and ABO (O is the circumcenter of triangle ABC),
> respectively, are concurrent.
> ...
> Does anyone know who is/was Kosnita ?
>
> Antreas