Contrary to what Felipe Pedrell indicates, the second Ave maris stella in his Victoria's collected works (vol. V, 1908, pp. 100-3, n° 33) doesn't appear in the collection published in 1600 in Madrid by the composer, nor in any other of the musician's books. In the 1600 edition, Victoria reissues the two first verses (plainchant followed by polyphony) of the Ave maris stella published in 1576 and then again in 1581.
The earliest source of the problematic Ave maris stella is Munich, Bayerische Staatsbibliothek, Musik-Abteilung, 2 Mus. pr. 23 handschriftlicher Beiband, dating from the third quarter oft he seventeenth century. This source is a manuscrit that runs as an appendix to the 1581 edition of Victoria's hymns. No attributions are given in the manuscript. The first attributions of the piece to Victoria arise in the nineteenth century, in manuscripts copied by Johann Michael Hauber, Johann Caspar Aiblinger, August Baumgartner and Carl Proske, and preserved in Munich and Regensburg. Proske pubished the piece in his Musica divina in 1859 (Annus primus, vol. III, pp. 419-24). The most probable hypothesis ist that Pedrell had knowledge of the second Ave maris stella, under the spanish composer's name, via Proske's Musica divina.
In all likelihood the piece is not by Victoria, not least because the composer has never written odd polyphonic verses of hymns. In his Studies in the Music of Tomás Luis de Victoria (2001), Eugene Casjen Cramer relies on the supposed authenticity of the work to ascribe the others pieces of Munich, Bayerische Staatsbibliothek, Musik-Abteilung, 2 Mus. pr. 23 handschriftlicher Beiband to the composer. These attributions should therefore be refuted.