GWRYGON GODDAU (standardized modern form)
Most of the lists of the children of Brychan Brycheiniog include a daughter Gwrygon, said to have married Cadrodd Calchfynydd, who is thought to be associated with the region around Roxburgh in what is now Scotland, placing her in our northern context. The dates calculated for her husband place him ca. 550, nearly a century later than the dates for the other two daughters of Brychan mentioned here. In general, the "core group" of Brychan's children fall in an early group and a late group, each of which could plausibly be contemporaries, but which are incompatible with each other.
Textual Sources
Gurycon Codheu - De Situ Brecheniauc (Cotton ms. Vespasian A xiv, fos. 10v-11v (ca. 1200) text perhaps a century earlierGrucon Guedu vxor Cradauc Calchuenit - Cognacio Brychan (Cotton MS. Domitian I, fox. 157v-158v) ms. dated 1502-55, copied from a ms poss of the 13th c.
[G]rugon - JC MS 20 written late 14th c., copied from ms ca. 1200
Gwrgon - Plant Brychan
Linguistic Analysis
Several of the sources for her name seem to confuse it somewhat with the masculine given name Gwrgan, and we should probably take the DSB form as the best base to work from. However the best linguistic connections for it cast some doubt on the historicity of this woman. Gwrygon is the Medieval Welsh evolution of the Romano-British place-name Viriconium (modern Wroxeter). The Old Welsh forms Guricon (Nennius) and Gureconn (in the poetry attributed to Llywarch Hen) are attested for the place. Jackson (LHEB p.601f) offers the Brittonic reconstruction *Uricono- for the place-name.
It is not at all unheard of for place-names to be transformed into personal names in the early Welsh tradition, either by inventing an eponym for which the place is said to have been named, or by mis-interpreting a reference to a place as a reference to a person. One of the most familiar examples of this is the creation of the name Myrddin (=Merlin) as a back-formation from the place-name Caerfyrddin, where we have clear evidence that the place-name derives from the Romano-British Moridunum "sea-fort". On the other hand, we also have rare examples where a name that clearly originates as a place-name somehow came into use as a personal name. A good example of this is the place-name Tegeingl, traceable to the British tribal name Deceangli, which Giraldus Cambrensis mentions as occurring as a woman's name (in fact, he offers a pun based on the name being both a woman's name and a place-name), and may also appear in the 1292 Lay Subsidy Roll for Skenefrith. So while co-existence as a place-name weakens the case for this as a valid personal name, it doesn't eliminate it entirely.
Following Jackson's derivation for the place-name, we can suggest a mid 6th century Latinized written form along the lines of Vricona, or following the earlier Latin forms of the place-name Viricona. This would correspond to a pronunciation along the lines of ['wri"-gon] or in English syllables "OORIH-gohn" but treating the first part as a single syllable, and where the "ih" vowel is more of a high schwa than the usual pronunciation of this sound.
I am uncertain enough about the possible derivation of the byname to hesitate to offer suggestions (since this is a side issue to the main thrust of the article). It may derive from the root cawdd (with variant codd depending on stress) with a meaning "anger, vexation, affliction", but while -au is a possible plural suffix, it is not the one normally found for this word.