There's a cool breeze coming through the window for the first time in forever, it seems, and all of a sudden I feel as though, perhaps, it won't be 998989 degrees in late August when I have to start teaching. (Last summer was pathetic in that way: We had better summer weather in early September than we did in actual summer.)

I've been practicing the guitar for the last little while, and have taken a break to jot down (if you write electronically, can you properly be said to "jot"?) some thoughts about guitar lessons while my fingers recover a bit . . . So far I've been practicing a very simple 'roll' (is it an arpeggio?) with my right hand. My initial thought is, "I paid $20 to learn this? I could have gotten a book!" Heck, I could get a book a week, as my loving spouse reminds me . . .

But still. I know the learning curve will be steep at first, but the hill eventually leads to more interesting places. I'm interested in how my little roll sounds so much better to me against a G chord than any other. When I'm playing a G chord, I hear the faint shimmerings of a tune that wants to come out. Some day, soon, I'll try to find it - but that would mean Deviating from the Roll, which is not what I want to do at the moment because I want to learn it first.

Last night, to provide myself with a little variety, I got down some of my song books and tried to play the roll as I changed chords and sang various songs. No luck whatsoever. For one thing, I have no idea how long the notes of the roll are supposed to last. Are they quarter-notes? Eighth notes? I realize those decisions are up to me - but I'm still slow enough that if I play songs, I tend to sound like a 78 record played at 33 . . . And I have no clue what to do if the chord changes in mid-measure. I only know how to do this if I start the picking pattern with my thumb.

And I was amazed to learn that in most cases, when you fret the strings to make the notes of a chord, the other, unfretted strings, are actually notes in the chord, too. I've been thinking a lot about it, and wanting to hunt up some chord charts to learn when it is and isn't true . . . It's a fascinating idea, wonderful in its symmetry.

But lest I get too excited about symmetry, I've been thinking a lot about octaves. Now, here's one for you Music Theorists out there. (I count at least four of you in my Known Readership.) If: An octave represents 2 notes distinguished because the frequency of their sound is double (or half, if you think of the higher note first), and you divide that octave into 8 notes (which should all theoretically be the same distance apart), why, then, do you end up with scales that incorporate flats and sharps? If, on a C major scale, the step between F and G is a whole note, why is it not also a whole note on the G scale? (Or on some scale that incorporates an F#.) This doesn't make sense to me.

Why is there no F flat?

I really want this all to make sense. I have the feeling that it won't, though, and that it's just the sort of thing I should have learned when I was 5, when most of the world didn't make sense and you just had to memorize stuff.

So somebody, please enlighten me. Or point me to a book that will. TIA . . .