Both text and notation of Ad caeli clara appear to have been written by the same scribe to judge by the ink colour and thickness of stroke.  The text is written in a clear minuscule with few ligatures and a few standard abbreviations.  Characteristic features include an occasional uncial ‘d’ with a wavy ascender, occasional curved descenders beneath the line for ‘f’ and ‘s’, and thickening at the head of several ascenders (especially on ‘l’, ‘b’ and ‘d’).  This combination of features implies a later eleventh- or early twelfth-century date of copying.

A letter written most likely in 1082 from Pope Gregory VII to Richard, Abbot of St-Victor at Marseilles, provides a terminus post quem for the addition of Ad caeli clara (for this letter, Epistola vagans 50, see H. E. J. Cowdrey, ed. and trans. The Epistolae Vagantes of Pope Gregory VII, Oxford: OUP, 1972, p. 120). Pope Gregory’s admonitory letter to his legate, copied here by another scribe, stands at the head of fol. 37r, after which Ad caeli clara continues.  Following the ‘Y’ strophe, the copying of Ad caeli clara is further interrupted by short alphabetical and other formulaic additions entered informally by several scribes of incomplete training.  These additions include a partial copying of the opening of the ‘X’ strophe and so were most likely added after the copying of strophes A-Y but before the continuation below.  The ‘Z’ strophe and doxology were copied beneath the additions, followed by a notated formula, Noeagis, and a notated incipit for Ad caeli clara.

Red ink was used for the the initials ‘A’ and ‘P’ and for the capital letters of each strophe repeated immediately to the left of the main writing space.  Red ink was also used for the title added at the foot of fol. 36r and for additions to the final two strophes, including line fillers, and a crossing out of the opening two words and final Amen of the doxology.  The additions in red ink appear to clarify the structure of the text in different ways.  The title specifies genre, indicating that the text is a hymnus of pentience or confession.  The addition of capitals to the left of the main writing space suggests an attempt to clarify the alphabetic pattern of the strophes, perhaps to aid legibility for singing.  The erasure of the beginning and end of the doxology implies an attempt to separate the doxology from the rest of the poem.

 Aquitanian notation was added to the ‘A’ strophe and its continuation into the ‘B’ strophe, as well as to the ‘P’ strophe and its continuation into the ‘Q’ strophe.  An empty line was left above the ‘A’ strophe and the ‘P’ strophe.  The resulting dry-point line above these strophes was used to regulate the disposition of the notation across the page; that is, the line was assigned a fixed pitch and signs across the page were heighted in relation to it.  Such a technique had been used by scribes writing Aquitanian notation from the late tenth century onwards; see, for example, Paris latin 1118, fols. 115 onwards, and latin 887, fols. 1-6 .  The practice of leaving empty lines for use in musical notation indicates that the text scribe anticipated the addition of neumes and increases the likelihood that text and notation were added by the same scribe.

 The notation consists almost entirely of single note signs drawn as a dash (tractulus).  The horizontal alignment of signs is sufficiently consistent to allow pitch height to be determined with confidence across the page.  The vertical alignement of signs is not always true (e.g., at ‘clara’), which suggests a degree of informality in presentation.  The tractulus is occasionally written at a diagonal angle, indicating the direction of the melody.  Other forms include the ‘m’-shaped Aquitanian oriscus, whose final tail on both appearances (at ‘redemptor’ and ‘genus’) implies a liquescence.  A further notable feature is that the notator consistently uses dots (puncta) ending in a tractulus when writing descending neumes: it remains to be determined whether this is a graphic trait, or an indication of either melodic or rhythmic profile.

 The Aquitanian notation used for the Noeagis formula maintains the puncta-tractulus manner of writing descending neumes and features two signs not seen until this point.  The first is an ‘r’ shape, which is the sign for a relatively high pitch or virga in other Aquitanian notations.  The second new sign is the longer hook shape written at a diagonal angle, which again appears to be a virga indicating a higher pitch.  The introduction of these virga signs may be due to the new requirement to indicate ascending pitches within a single neume, as well as a desire to indicate as much as possible about relative pitch height within the Noeagis formula.

The melody for Ad caeli clara may be securely reconstructed due to the consistent heighting of neumes around the dry-point line and the Noeagis formula.  One of the apparent functions of Noeagis formulae was to establish the mode of the following chant; in this case, the formula matches that used for mode 4, otherwise known as the deuterus plagal or plagal mode on E.  The transcription given below aligns the Noeagis formula in this manuscript with one for mode 4 transmitted in Paris lat. 909, whose Tonary was copied in the first third of the eleventh century. The partial erasure mid-way through the Noeagis formula appears to result from confusion over the placement of the letter ‘e’: the notator seems originally to have ignored the change of vowel, then on realising the mistake erased the first attempt and correctly isolated the pitch F with its own neume.  Noeagis formulae do not normally proceed beyond the final of the mode, in this case the note E.  Continuation of the formula here appears to be motivated by a desire to clarify the pitches of the incipit of Ad caeli clara (Adcaeli_Pa32.1). 

With the mode and pitches of the incipit established, the melody for Ad caeli clara as notated on fol. 36v can be securely reconstructed once it is understood that the pitch assigned to the dry-point line changes: it is E for the ‘A’ and ‘B’ strophes, then G followed by F for the ‘P’ and ‘Q’ strophes.  The resulting melody could be classified in either mode 3 or mode 4 on the basis of its range; it was probably assigned to mode 4 because of the emphasis on a as a recitation tone (Adcaeli_Pa32.2).

The melody could be said to follow an AA’BC structure since the melodic profile of the opening line is repeated in the second with an extension up to the high c at ‘levare meos’, whose pitches in brackets are confirmed by the notation added to the ‘P’ strophe.  The third line seems initially to have a different profile as it remains on a as a recitation tone at ‘depressus’; however, the melody of the third line can also be heard as delaying the second line’s rise to the high c until ‘peccatorum’ followed by a truncated continuation, ending with the first open cadence on a.  The final line emphasizes notes either side of the final before settling on the final E. The shape of the melody is particularly suited to the semantic content of the opening strophe, reaching the high c for the first time at ‘levare meos’, refusing to rise at ‘depressus’, and cadencing with an extended emphasis on lower tones at the closing invocation, ‘parce redemptor’.  Turning to general features, the melody makes use of the so-called Gallican cadence (here DEE and Gaa) as commonly found in early medieval sequences, and the final E is offset by repeated emphasis on the notes of a clearly defined third chain (D-F-a-c). 

The melody reconstructed for the ‘A’ strophe is identical to that recorded for the ‘P’ strophe and the continuations into the ‘B’ and ‘Q’ strophes.  It does not match any other recorded for Ad caeli clara and also lacks the pattern of melodic organization shared between several versions notated in earlier manuscripts.  The use of the Noeagis formula and its link to the incipit of Ad caeli clara also implies a need to clarify the mode of a unfamiliar melody.  Although monophonic melodies with E finals are relatively rare in the Aquitanian song repertory recorded from the late eleventh century onwards, melodies with E finals are found in northern French song repertories of the eleventh century: see, for example, the melody for Gottschalk’s O mi custos recorded in Paris lat. 1928. 

The question that remains is why strophe ‘P’ was set apart for a new notational entry.  The most likely explanation is that Ad caeli clara was divided into separate songs sung to the same melody; that is, the opening 14 strophes (A-O) were performed as one song, and the following 9 strophes (P-Z) as another; both might have made use of the closing doxology.  Precedent for this practice is found in the hymn repertory, in which longer hymns were often divided into shorter versions for performance in the Office.  If Ad caeli clara was divided in two in performance, then beginning anew at the initial imperative ‘Plangite’ would mirror the wider practice of opening songs with an injunction to join in singing or lamentation.