I'm not as enthusiastic as the previous reviewer. I have taught music theory at all levels for 17 years, and while there is good material in Spencer's The Practice of Harmony, there is much that seems downright peculiar. If your theory professor assigns this book, of course you've got to use it; otherwise, much of the material is presented more clearly in other sources. The problems are twofold: reliance on standard theoretical concepts divorced from practice or context, and Spencer's attempt to either explain concepts in new ways or offer shortcuts in learning or remembering the concepts--it's a laudable effort, to be sure, but one which leads him astray.
Take, for example, this wonderfully abstruse line from his "Suggestions and Strategies" for learning triad quality: "If the root of a triad is natural and the fifth is flat, or if the root is sharp and the fifth is natural, or if the fifth is double-flat, the triad is diminished." (p. 69) All that may well be true, but if you've memorised that and still can't tell that the third of the chord is minor and the fifth is diminished, you have no recognition of what a diminished triad IS. There's quite a difference between naming and understanding, and Spencer ignores this.
Another and much more significant problem occurs with Spencer's explanation of the minor. Frankly, he doesn't seem to understand that there is not a "harmonic minor key" or a "natural minor key"--he does not use the terms, but in the section "Triads in Major and Minor Keys" (p. 68) he shows the triads first on C major, then on A minor (natural), a minor (harmonic), and a minor (melodic)--ascending only. Now, while there are SCALES in these various forms of the minor, they represent quite different things from the triads that are found in most common practice period music. Even triads built on the harmonic minor scale, with a triad of III+ (augmented) do not reflect the actual triads in most music--III is much more commonly found in the music as a major triad than as an augmented one, regardless of the theory. Minor is a complex situation, not easily explained; but a piece is in a minor mode which uses the various forms of minor--melodic for the melodies (mostly), harmonic for the harmonies (mostly, except III), and in Classical music, almost never natural--purely a theoretical construct because of the key signature linkage with the relative major. Natural minor in Classical music occurs as the descending form of the melodic minor. The real problem is that the minor scales are all theoretical constructs and not reflective of the music--the issue is how to handle scale degrees 6 and 7. To solve this, rather than address it directly, Spencer introduces yet another theoretical construct, the 'synthetic' minor scale, made up of a combination of all the other forms together. It's an appalling idea, one I've never seen elsewhere, and totally without context. Spencer would have been better served by introducing the minor mode in its own chapter, rather than trying to tack it on as an appendage to the major mode.
Lack of context lies at the heart of Spencer's approach, and to be fair this is true of far too many theory books. Instead of discussing the basis of functional harmony in the Classical Period, for example, we get the same tired "Primary Triads in Root Position". I'm of the firm belief that discussion function of chords--Dominant preparation for ii and IV, Dominant function for V and vii, Tonic for I and Tonic substitute for vi in major--is much more useful than listing all the "rules" of doublings, etc. Similarly, cadence types are named, but the functions of the various types are not described.
All in all, Spencer's approach in The Practice of Harmony is an odd combination of pretty standard pedantic theory and new, sometimes confusing approaches. Theory is really nothing by itself; without a musical context it is meaningless. Yet Spencer never attempts to provide such a context; indeed, he uses not a single example from actual music literature. You are better served elsewhere.