In my field (SA land and agrarian studies, though I'm historical) Andries du Toit is quite renowned and a very careful researcher. His facts are on point--the constituencies in rural areas did represent fewer people per seat than urban areas, and in SW Africa (added after their majority) it was fewer still--think 8000 per seat instead of 10000, with Coloured voters concentrated in those urban spaces not counted in these totals at all (and black voters not even existing on the national rolls since 1936). They did do an end-run via Senate to wipe out the Coloured vote, and later engaged in other 'corrections' of policy and law to address oversights. [edit: It's worth pointing out too that the Nats didn't expect to win in '48 and had little real idea of what to do when they did. They took measures step by small step, closing off the options for reversal and entrenching themselves with an expanded Afrikaner bureaucracy to run all this new policy machinery--packed with Party faithful, meant to assure that any loss of power would be temporary and blunted.]

It's harder to speak to the sense of how it felt to liberal Afrikaners but it's not unthinkable--Hermann Giliomee (The Afrikaners, 2nd ed) may be a good second source for that. The SA liberal wing in general was in disarray following the loss, and they didn't have much of an ascendancy in the United Party anyway, which was also racist but not committed to such a simple buzzword ideal and tied inextricably to Afrikaner nationalism. The sense that some moderation would take over really only disappeared with the full ascendancy of Hendrik Verwoerd (even before he was PM--under Strijdom especially) and the institutionalization of a pseudo-scientific basis for religio-nationalist racialism that, in reality, manifested in racial supremacy. Some white liberals stated later that they always saw this, but I can't discuss that without anecdotes sadly. Glenn Moss, in his memoir recently published (the title escapes me, argh), describes coming of age in the 1970s and how often white liberals still tended to 'tsk tsk, ag shame' and look the other way, not unlike their counterparts elsewhere.

Saul Dubow's done the really key work on the rise of 'scientific' racism (Illicit Union, which has different title here in the US) and that understanding is baked into his 2014 general history of apartheid which is the standard volume now) but the one Du Toit recommends for the 1940s (also co-edited by Dubow and labour historian Alan Jeeves) also has good buzz. It was a moment of possibilities--but also deeply reactionary elements. Yes, when stonewalled, they did bend the rules to the point of breakage, in exactly the manner he describes.

As for comparative works in settler colonial societies, there's an entire journal (Safundi) dedicated to US/SA comparative studies; there's also a journal, as you likely know, devoted to settler colonialism (Settler Colonial Studies). But in terms of directly comparative volumes we need an updating of George Fredrickson's two key US/SA texts, White Supremacy and Black Liberation. They're over a generation old now. Other comparative works exist but I'll have to see if there's anything new that is widely comparative using a settler-colonial framing. I'm not sure who is doing that kind of work specifically in the theme you describe, but you may find some things/cites in Settler Colonial Studies or the work of Lorenzo Veracini, like him or not. The state of the field is in some flux as new framings emerge for further research and writing.

[edit 2: readability]