I was afraid someone would say that
It’s a pretty tall order!
This is an interesting book on the
restoration of Gregorian chant in the 19th century:
http://www.solesmes.com/node/9132?position=2&list=3q4NLWuXQVUUzhnCoSOa-yta9jcwTN_gCwQhYS1G9sg
Alas it appears to be out of print.
That said, I think it’s a pretty fair assessment that Gregorian chant as we know it, dates from the late 19th century and was made official by the Vatican Edition of the Roman Gradual promulgated by Pius X in 1908. Now, that’s a facile and somewhat facetious answer. The restoration of chant by Solesmes is somewhat controversial because of some musical signs added in the process (that have been removed in the most recent antiphonary for the Monastic Divine Office) and the fact that the modern chant repertoire as codified in the chant books they produce are specifically *Solesmes’ interpretation of Gregorian chant" and thus it is how Solesmes imagined chant sounded. How it actually sounded, is anybody’s guess.
The corpus of music that Solesmes used in its restaurations, from actual antiphonaries (St. Gall, Laon), historians generally agree was written in the 8th to 10th centuries. Some may extend that to the 12th or 13th. There is no definitive answer. Anything later, such as the Mass of the Angels (Mass VIII), can be considered “Neo-Gregorian” IMHO. Much of the Mass of the Angels was written in the 15th and 16th centuries (except the Sanctus), and Credo III often used with the Mass VIII setting is from the 17th. Some chant is downright modern: the propers for the feast of Christ the King, were written in the 20th century, since that is when the feast was introduced.
Gregorian chant, that which we know from existing manuscripts, is from the Carolingian era (8th to 10th centuries, and is a fusion of Old Roman and Gallican chant. It is from these manuscripts, for the most part, that Solesmes restored the chant we use today. There are no known manuscripts of Gallican chant known to survive.
There have been some predecessor and parallel developments, but alas their specific histories are lost in the mists of time, such as Ambrosian (Milan), Gallican (France), Mozarabic (Toledo, Spain), Sarum (England), and Beneventan (Southern Italy).
As my focus is largely performance (of modern Gregorian chant!) and not history, I don’t claim my account to be authoritative, and I believe that Solesmes’ account linked to above focuses more on the history of the restoration of the Gregorian repertory and not so much on the history of the sources.
For more scholarly writings on the subject, I would recommend consulting the reference section of the Wikipedia article on Gregorian chant, you’ll find lots of scholarly works to search for.