It is evident, therefore, that several architects were employed upon the work, and I know few facts in the history of mediæval art more interesting than the account we have here of the payment of an architect whilst he travelled to find some good work to copy for the city of Valencia. The steeple of Lérida cathedral will be mentioned in its place, and it is sufficient now to say that it is also octagonal, of great height, and dates from the commencement of the fourteenth century. I know nothing at Narbonne which could have been suggestive to Pedro Balaguer, but the city was Spanish in those days, and is probably only mentioned as one of the most important places to which he went.

When the Micalete was built the nave of the church seems to have been still unfinished, the choir and transepts and part of the nave only having been built. In 1459, under the direction of an architect named Valdomar, a native of Valencia, the work was continued, and the church was joined to the tower. The authority for this statement is a MS. in the library of the convent of San Domingo, Valencia, which says: “In the year of our Lord 1459, on Monday, the 10th of September, they commenced digging to make the doorway and arcade of the cathedral; Master Valdomar was the master of the works, a native of the said city of Valencia.”[259] Of Valdomar’s work in this part of the church nothing remains, the whole has been altered in the most cruel way, and the most contemptible work erected in its place. Valdomar appears to have died whilst his work was in progress, and to have been succeeded by Pedro Compte, who concluded the work in 1482. The manuscript already quoted from the library of San Domingo is the authority for this statement, and describes Pedro Compte as “Molt sabut en l’art de la pedra.”[260]

On the south side of the nave there is a Chapter-house, which is said by Ponz[261] to be the work of Pedro Compte, and to have been built at the cost of Bishop D. Vidal Blaues, in A.D. 1358. If this statement is correct, it follows that there were two architects of this name, the second having erected the Lonja de la Sedia, to which I shall have presently to refer, in A.D. 1482. The tracery of the windows, and the details generally of the Chapter-house, is so geometrical and good, that it is probable that the date given by Ponz may be depended upon. It is a square room nearly sixty feet in diameter, and groined in stone. The vault is similar to those which I first saw at Burgos, having arches thrown across the angles to bring it to an octagon, and the triangular compartments in the angles having their vaults below the main vault. It is lighted by small windows very high up in the walls on the cardinal sides, and these are circular and spherical triangles in outline, filled with geometrical tracery. On the south side is a very elaborate arcaded reredos and altar, and on the west a pulpit corbelled out from the wall. The design and detail of the whole are extremely fine, and I regret that I was able to make but a very hurried examination of it, and no sketches; meeting here, almost for the first time in Spain, with a sacristan who refused to allow me to do more than look, the fact being that it was his time for dinner and siesta!

In the old sacristy to the east of this room are still preserved two embroidered altar frontals, said to have been brought from our own old St. Paul’s by two merchants, Andres and Pedro de Medina, just about the time of the Reformation.[262] They are therefore of especial interest to an Englishman. They are very large works, strained on frames, and were, I believe, hangings rather than altar frontals, as they are evidently continuations one of the other. The field is of gold, diapered, and upon this a succession of subjects is embroidered. On one cloth are (beginning at the left) (1) our Lord bearing his Cross; (2) being nailed to the Cross; (3) crucified, with the thieves on either side; (4) descending from the Cross; (5) entombed. The next cloth has (1) the descent into Hell; (2) the Maries going to the sepulchre; (3) the Maries at the tomb, the angel, and (4) the Resurrection. The effect of the whole work is like that of a brilliant German painting, and the figures are full of action and spirit, and have a great deal of expression in their faces. The diapered ground is made with gold thread, laid down in vertical lines, and then diapered with diagonal lines of fine bullion stitched down over it to form the diaper. The gold is generally manufactured in a double twist, and borders and edgings are all done with a very bold twisted gold cord. The faces are all wrought in silk, and some of the dresses are of silk, lined all over with gold. The old border at the edge exists on one only of the frontals. The size of each is 3 ft. 1 in. by 10 ft. 2 in., and the date, as nearly as I can judge, must be about A.D. 1450. There is also preserved here a missal which once belonged to Westminster Abbey.

I could find no other church of any interest. There are several which have some old remains, but they are generally so damaged and decayed, that it is impossible to make anything of them. One I saw desecrated and occupied by the military, and was unable to enter; and there is another in a street leading out of the Calle de Caballeros, which has a very fine round-arched doorway, with three shafts in the jambs, and good thirteenth-century mouldings in the arch, and which is evidently of the same age as the south door of the cathedral. The capitals have each two wyverns fighting, and the abaci are well carved. The church, however, was desecrated, and no one knew how I could gain admission to it.

The walls and gates are of more interest. They are lofty, and generally well preserved. The two finest gates are the Puerta de Serranos, and that del Cuarte. The former, said by Ford[263] to have been built in A.D. 1349, is a noble erection. Two grand polygonal towers flank the entrance archway, which is recessed in the centre. Above this the wall is covered with tracery panelling, and then a great projecting gallery or platform, supported on enormous corbels, is carried all round the three exposed sides of the gateway. The towers are carried up a considerable height above this gallery, and it is probable that there was originally a wooden construction over it, of the kind which M. Viollet le Duc, in his treatise on military architecture, has shown to have been commonly adopted in fortifications of this age. The Puerta del Cuarte is of the same description, and has two circular flanking towers, but is less imposing, and is said to have been built in A.D. 1444. Both gateways are completely open at the back, enormous open arches, one above the other, rendering them useless for attack against the city; and the corbelled-out passages at the top are not continued across the back.

The domestic remains here are of some importance. One feature of rather frequent occurrence is the window of two or three lights, divided by detached shafts. The earlier examples have simple trefoil heads, and sculptured capitals to the columns. In the later examples there are mouldings round the cusped head, and the abaci and capitals are carved: but it is a very curious fact, that wherever I saw any old towns on the coast of the Mediterranean, there I always saw some specimens of this later kind of window, with detail and carving so identical in character, that I was almost driven to the conclusion that they were all executed in the same place, and sent about the country to be fixed! Nevertheless, they are always very pretty, so that one ought not to grumble if they do occur a little too often. The shafts are generally of marble, and often coupled one behind the other.

The Arabs had a name for this class of windows, and as we have not, and want one, it may be as well to mention it. They are called ajimez, literally windows by which the sun enters. The Arabs seem to have supplied many of the architectural terms in use in Spain, and probably we owe them in this case not only the name, but the design also. Among other Arab words still in common use, I may mention Alcazar, Alcalá, Tapia, and many more are given in vocabularies.