Late Roman Army Grade/Rank List under Anastasius

[Longovicium]

Hyginus's camp holds three legions, I think, and most likely dates to the early 2nd century - I can't remember who suggested that the 'Pannonian Veredarii' might be cavalry detached from the three legions (which may indeed have come from Pannonia), but if so we would have around 260-270 per legion - that does look rather Perge-like, but the legions in question were still c.6000 men so we shouldn't go too crazy with that idea!

I do suspect that these 'veredarii' cavalry at Perge were somehow a substitute for the old legion cavalry that had been detached as Promoti.
Meanwhile, a little more on veredarii in earlier military formations (one of these inscriptions has already been mentioned by Marcel):
CIL 03, 13795 (Rakovitza, Dacia, AD138): n(umerus) burg(ariorum) et veredario(rum) Daciae inf(erioris) sub / Fl(avio) Constante proc(uratore) Aug(usti)
CIL 13, 07439 (Kapersburg, Germania): In h(onorem) d(omus) d(ivinae) Genio / veredariorum n(umeri) N(idensium)
CIL 13, 08492 (Koln, Germania): ...nus vered(arius) [et] / [3]stis Dirmes[us(?)] / [vetera]nus item cu[rat(or)] / [n(umeri) Britto]num cum
All three appear to show veredarii attached to frontier numeri in the 2nd and 3rd centuries. As Marcel noted, the Numerus Nidensium one included a dedication to the celtic horse goddess Epona.

 

Ah nice find re the veredariorum/ius - thanks for sharing that. I did wonder what epigraphical evidence for veredarii existed outside the Perge Edict. 

As for the Promoti, Pavlovic's thesis raises some interesting points: he states that the promoti as we known them in the ND were legionary cavalry brigaded away from their legion and which never returned to the province where the legion was billeted and hence its administrative umbrella, as it were. In other words, any legionary cavalry which were detached and despatched beyond the administrative reach of the parent legion were reconstituted as stand alone cavalry units - the Equites Promoti listed in the ND. However, he suggests that promoti who were detached and/or brigaded with others but who subsequently returned or never moved outside the provincial zone of the parent legion remained 'on the books' and would therefore not appear in the ND as separate stand-alone units. The general theory that all legionary cavalry were hived-off sort of falls apart here if he is correct. This makes persuasive reading and would account for the veredarii listed with the Perge legion.

But there's more: Pavlovic argues that, following a careful reading of Vegetius, there were 2 classes of legionary cavalry listed on the centurial rolls. The main body attached to the cohorts apart from the First Cohort were all what Vegetius lists as 'contati' units however Vegetius distinguishes them from another legionary cavalry type called the 'loricati'  or armoured. Given that the mainstay cavalry were armoured anyway, he thinks this refers to heavy armour or cataphract cavalry (Vegetius 2.16). He states on page 114 of the thesis that the legionary cavalry equalled 4 turmae for the First (double-sized) cohort and 2 turmae for each of the remaining nine cohorts. But these turmae were not necessarily proper turmae merely tactical units. This is an interesting division for it implies a ratio of heavy cavalry to lance or contus cavalry which might mirror the Perge Edict. I am thinking of the two grades of veredarii here and wonder if the 50 veredarii and the 225 veredarii alii might in some manner echo this distinction?

Maths is not my strong suit, however . . .