ABSTRACT

These lines in ink are barely legible because of the damage to Nbk 20 through its immersion with the Don Juan in July 1822. Salvaged in September of that year, this nbk was repaired in the early 1920s and, as noted in Shelley’s Guitar 158, the surviving leaves of the original are often less legible than the two sets of photographs made of them immediately after restoration, now in the Bodleian Library (MS. Shelley adds. e. 5 and MS. Shelley adds. e. 27). The approximate date range of the entries in the surviving leaves of Nbk 20 appears to be February–June 1821. The versos of the leaves that precede this fragment comprise part of S.’s intermediate fair copy of DP, the rectos (mainly) drafts of the Preface to Adonais (no. 403) and To Emilia Viviani (no. 393). This fragment is drafted in the upper half of f. 6r. Beneath, and apparently entered afterwards, is a cancelled phrase in darker ink, to testify what base which, as Donald H. Reiman suggests (BSM vii 339), may relate to the drafts of the Preface to Adonais, and was probably drafted towards the end of the composition period of that poem (c. 19 May–c. 16 June 1821) – as explained in its headnote. Although several words are impossible to decipher with confidence, these lines appear to be in terza rima and it is conceivable that they relate to S.’s drafts of terza rima in Italian on ff. 24–7 of this nbk (Dal spiro della tua mente, [è] istinta (no. 395)). The third line of that fragment – La guancia dal cadente sole tinta (‘The cheek tinted by the setting sun’) – has parallels with the twilight referred to in the second and third of the lines below. It is possible therefore that these lines were drafted before S.’s experiment with terza rima in Italian a few pages later. It may be inferred from the proximity of these lines to Dal spiro della tua mente, [è] istinta and To Emilia Viviani in Nbk 20 that their context was S.’s 195increasing interest in Teresa Viviani in the early months of 1821. A date of composition in February or March 1821 therefore seems most likely.