I don't intend to write your friend's paper for him (it's against the rules, for one thing, besides you know...academic integrity), but I'm afraid he's running a doomed race here. The existence of the so-called "Marian Reforms" has been questioned at best or, more usually, rejected outright by pretty much every scholar of the Republic precisely because of a near-total lack of evidence. The very concept appears to have arisen from the observation that there is a major discrepancy between the army described by Polybius and the army described by Caesar, when we next get good information about the army's structure and composition. There's another gap in our information between Caesar and Augustus, under whom we have surprisingly good information. Either sometime during the Victorian Period or the early 20th century some scholars attempted to identify this change as occurring at the very end of the second century with Marius' first election to the consulship, on the grounds that Marius supposedly abolished the property qualification for recruitment into the army, which was seen as the crucial step into creating a professionalized "client army." Unfortunately, none of those things appears supported in the texts, and the theory of the "client army" has been pretty resoundingly rejected. The sole independent attestation to this supposed abolition of the property requirement is from Sallust, Iug. 86:

Ipse interea milites scribere, non more maiorum neque ex classibus, sed uti cuiusque lubido erat, capite censos plerosque

Meanwhile he [Marius] enrolled soldiers not, according to the mos maiorum, according to the property classes, but according to the desire of the recruits, many of whom were from the capite censi

There are a handful of later references to Marius' enlistment of the capite censi (I think two), but they seem to be derived from Sallust and contribute nothing further. But what Sallust says is not that Marius abolished the property qualification, but that at a time of perceived emergency (Sallust cites the likelihood that he could not find landed recruits in time) Marius enlisted the unlanded. The procedure was hardly unknown: in 215 Sempronius Gracchus enlisted a slave army. Nor is there any evidence for such a change in recruitment. Actual patterns of recruitment cluster around rural regions, particularly those parts of Italy known in antiquity for their martial traditions, and only in times of emergency do we find levies in the city. The second part of the so-called "Marian Reforms," more cited by military historians than other scholars of the period, is the supposed change in equipment under Marius. But there's only a single reference to such a thing, from Plutarch, Mar., 25, where he describes Marius' supposed change in the design of the pila. But there's no evidence to suggest that Marius did any such thing, and it's unknown when such a design came into circulation, if it ever did--the use of the pila that Plutarch describes seems incompatible with what he says about Caesar's troops, and what Caesar says himself about the pila.

The evidence for the two supposedly great changes in the army under Marius is therefore nearly non-existent, and without them the whole concept starts to fall apart. And as Keaveney and Morstein-Marx have shown pretty conclusively the evidence that "client armies" followed their generals mindlessly is flimsy and ignores all the rhetoric we see surrounding the "defection" of these armies, particularly in 88 and 87 when the "client army" is supposed to have appeared first in a civil war. Indeed, one of the armies (Cinna's in 87) could not possibly have operated according to this model, since the army wasn't even Cinna's, but App. Claudius'!