I just finished the book Romance Languages - A Historical Introduction (RL), and want to just say, "Wow." This was a superb book, and I take great pleasure in recommending it. Specifically, for someone who likes language(s), and finds the processes that drive sound change and morphology interesting, this is an excellent read.
From the start, RL singles out five languages to base its comparative studies on, French, Italian, Portuguese, Romanian and Spanish. There are in fact other Romance languages in existence, and the distinction (or lack thereof) between a dialect and language plays a part in the discussion. But these five are chosen because they have gone on to become national languages. That being said, Romanian and Portuguese are treated sparingly through the bulk of the text, Romanian more so than Portuguese. However, each of these receives its own chapter to make up for this, owing to the distinctive historical trajectory each language takes, differentiating them from the main stream of Romance and Popular Latin. The Romanian chapter leaves quite a bit open for discussion. It also makes a number of tentative claims, which is expected since historical reconstruction is not at all clear for this language.
The Popular Latin presented in this book was given quite a bit of support, and is essential for understanding the changes that drove sound change from Latin towards the distinctive modern Romance languages. However, since much of this transmission was oral, the book is continually having to tag word forms with an asterisk, indicating that there is no direct textual evidence of the form. Obviously one should read this text and understand that there is still a lot of study in this field, with competing opinions about many of the historical changes in specific forms (and the authors are careful to make this clear).
This book is aimed at students with some linguistic study under their belt. It is not as dense as some texts I have seen that demonstrate the phonology and morphology of a language, nor is it an easy read for someone without a linguistic background. Prior study of Latin is expected (I am self-taught), and some prior study of linguistics is helpful, as it relies heavily on IPA to describe forms. I don't have as good a grasp on IPA as I would like, but I have studied many different languages and thus have covered articulation and phonetics more contextually. I found this book challenging, without being overwhelming. In fact, by the middle, I had a much better handle on IPA symbols that relate to the Romance family. Learning in context is a powerful tool, without doubt.
As phonetic and morphological change processes are named, they are given good descriptions (and each of these terms is included in a guide near the back). This was very helpful. Not as much time is spent on terminology for articulation, and this was a little bit more problematic as it is an important concept for dealing with sound changes. However, RL tended to describe changes in terms of the processes that occurred rather than the physical articulation that allowed or promoted those changes, so the emphasis on process vocabulary makes sense.
The book is littered with practice exercises and inline questions to force grappling with the information. The end of each chapter provides additional questions that require more analysis and creative thought about the processes that have been looked at. These were excellent, and make this text more than just descriptive. However, the major failure of this book, in my mind, is there is no answer key. Other than the inline questions, no answers are provided for practice exercises or chapter-end exercises. This made the exercises interesting, but something short of what might have been hoped for.
Finally, the bibliography at the end is extensive and will allow the interested reader to delve where they will. As someone who read this on his own time (not as part of a linguistics course or program), that is very helpful.