In 1842 Berlioz was certainly among the very earliest to mention the «saxophone» in his famous comment in the Journal des Debats. His text is often cited in historic sketches on the saxophone. It is time to put a question mark in margin of this text, or an exclamation mark, or maybe both.
His comment dates from almost four years before the patent on the saxophone was granted. There have been writers previous to Berlioz to mention the saxophone, or an instrument which we now think might have been a saxophone prototype, but as these reports are not very detailed technically, they are not relevant to the present subject. The instrument that Berlioz had heard must have been such a prototype or even the same prototype and, as we will see, this instrument must have differed in quite an interesting aspect from the saxophone as built later on by Adolphe Sax. Berlioz writes:
The saxophone, named after its inventor, is a brass instrument with nineteen keys, whose shape is rather similar to that of the ophicleide. Its mouthpiece, unlike those of most brass instruments, is similar to the mouthpiece of the bass clarinet. Thus, the saxophone becomes the head of a new group: brass instruments played with a reed. It has a compass of three octaves beginning from low-Bb under the bass clef, its fingering is similar to that of the flute or the second register of the clarinet. Its sonority is of such a nature that I do not know of any low instrument presently in use that can compare to it. It is full, mellow, vibrant, with an enormous power and easy to play sweetly. It is much superior in my mind to the low notes of the ophicleide in its tuning and flexibility; moreover it is completely new and does not resemble any of the timbres one hears in the present orchestra, with the sole exception of the bass clarinet's low E and F. Because it uses a reed, the saxophone can increase or diminish its sound. In its upper register it produces notes of a penetrating resonance that can be successfully applied to melodic, expressive lines. Without doubt, it will never be appropriate for rapid passages or for complicated arpeggios; but low instruments were not meant for nimble movements. Instead of complaining, we must rejoice that it is impossible to abuse the saxophone and destroy its majestic character by giving it mere musical fripperies to perlorm. Composers will be very indebted to Mr. Sax when his new instruments are generally employed. If he perseveres, he will meet with the support of all friends of music.
Journal des débats politiques et litéraires, June 12, 1842, translation as given by Albert Rice.
I highlight both the 19 keys and the low-Bb. Such an instrument is not known to us. The saxophone, as it was described in fair detail in the 1846 patent, is, what we now would call, a baritone saxophone in Eb. It was equipped with 20 keys, among which were two register keys. Its compass ranged from a [written] low-B up to a [written] high F. In the patent Sax strictly numbers holes and the corresponding keys from the bottom up, starting with 1 for the key for low B–C and ending with 20 for the upper register key. The lower register key is numbered 16 (by the way, we may note that later on this vent has shifted a step or two upward). The alto and the tenor saxophones as subsequently made by Sax indeed both have 18 + 2 keys, giving them the said compass, whereas the baritone and the soprano only had 16 + 2 keys, giving a compass from low-B to high-Eb. Both are even numbers of keys. We won't bother too much about Sax describing a baritone going up to high-F, which he then never took into production.
What is really intrigueing though, is that Berlioz mentions 19 keys, an odd number! We might interpret the 19 as either meaning 19 + 2 (counting the register keys separately) or as 17 + 2 (including them). Given that Sax in the F3226 patent numbered his register keys right among the tone hole keys, the 17 + 2 seems to be the most logical option. Berlioz also mentions the instrument going down to a low-Bb. What is not well defined in the text is whether or not it is about a transposing instrument. If it was, was it then in Bb going down to written C, or was Berlioz talking transposed notes – an instrument in indefinite key going down to a written Bb? And if it wasn't, was it then an instrument in C going down to Bb? Seeing the text of the 1846 patent, the latter two options are very well possible; an instrument going down no lower than to a written C is less likely. Anyway, with 17 tone holes and 2 register vents, it would mean that this instrument would then go up to the [either written or sounding] high-Eb, such as was indeed usual for the baritone saxophone later on. The odd number 19 is then consistent with the low-Bb Berlioz mentions.
Whatever the case might be, such an instrument is not known to us. Now we can either choose between the composer Berlioz not knowing to distinguish between a B and a Bb and not being able to count beyond 10 or going by hearsay alone, or us recognizing the existence of a prototype going down to a [written] low-Bb – an instrument now lost to us and a lowest note for the saxophone which Adolphe Sax would never ever make again. It would take about half a century for the low-Bb to return and then in a saxophone that had meanwhile changed shape.
We can only speculate as to the reason why Sax dropped the low-Bb, but at his point, I would surmise that it could have to do with acoustical stability. What leads me to this suspicion is my own attempt in building an alto saxophone with a parabolic bore (see «A TRIBUTE TO SAX», THE MAKING OF THE BORE). It turns out that, while a low-C is still no problem at all, it is difficult to make it go down to a low-Bb in a stable fashion. The instrument tends to produce involuntary multiphonics. At the same time, as Berlioz' text is so well known, we might be surprised to see that the 19 keys appear to have escaped attention so easily.