and variety of the works included. Indeed, the can be regarded as a kind of retrospective visual diary or commonplace book that provides a mini-overview of Rossetti's development as an artist.

Part I includes at least six drawings that can with certainty be regarded as the original conceptional sketches for major pictures, similar to my study for (S.Appendix 4, not reprinted in the ), which was given such a prominent position in the catalogue of the 1973 Rossetti retrospective exhibition at the Royal Academy. In this class are the studies for (1), (9), (12a), (13), (14), and (17). Another six— (2), — (3), (11), (15), (16), and (20)—if they are not the initial studies for the pictures in question, clearly represent important stages in the evolution of the final conception.

Studies for executed pictures, because they lend themselves to comparison with the finished works, have an inherent interest for students of art. Studies for unexecuted pictures, even when they are only visual fragments like some of those in Part II, provide valuable insights into the workings of the artist's imagination. When, like (24), (25), (30), the (32), and the (33), they also exhibit a higher degree of finish or execution, they transcend any taxonomical attempt to relegate them to a subordinate class and assume independent status in the canon.

In presenting the drawings, which have no particular order in the portfolio, it has seemed expedient to arrange them in descending order of interest rather than to follow a strict chronological pattern along the lines of Surtees and most catalogues raisonné, even though inevitably this plan necessitated some concessions and arbitrary decisions. The juvenilia in Part VI, for example, while important iconographically, are in the main less visually attractive than works consigned to other sections. A clear exception, however, is the last drawing in Part VI, (113). Among the best of Rossetti's early drawings in the , it is placed here rather than among the Early Drawings in Part III, which in general consists of more finished works, only because of its probable date. In Part IV, the portraits and better caricatures in the are grouped irrespective of date by size or subject, without regard to whether the drawing technically belongs among the juvenilia. Some pairings of drawings on a single plate have, admittedly, been dictated by spatial considerations, but throughout the emphasis has been on achieving aesthetically satisfying pages through the rational grouping of related drawings.

The unimpeachable provenance of the precludes any question of attribution. Virtually all the drawings are annotated by William Michael Rossetti, whose archival instincts led him to preserve and record, both for posterity and for his own use in his many publications on his poet-artist brother, all documents relating to Dante Gabriel's literary and artistic productions. Because William's familiarity with his brother's work was so intimate that even his speculations are normally reliable, his annotations are cited in full in the Catalogue. That he was not infallible, however, is indicated by his misidentification of Plate 11 as a study for rather than for ; and two of the most amusing annotations on the drawings are exchanges, written years apart, between William and his daughter Helen Rossetti Angeli, in which she challenges her father's reservations about attributing a particular drawing to her uncle (see 58, 101). William's dating is frequently tentative, but more often than not accurate, at least within the range he suggests. One of the drawings he identifies as his own (the reverse show-through portrait of Holman Hunt in 50a); another appears to be by Maria Rossetti (the juvenile sketch of Clifford from Henry VI in 99a). The remainder are all by Dante Gabriel.

Needless to say, the annotations in the Catalogue rely heavily on William's notes on