Although there is not any news about the construction of this church, we find already a reference in a donation by Alphonse the Third in 905, which coincides with the dating that can be made based on its main characteristics, that locate it in the kingdom of Alphonse the Second. As with so many other Asturian Pre-Romanesque churches, it was destroyed in 1936. It was not discovered until 1954 by J. Manzanares and rebuilt later in 1958 by Luis Menéndez Pidal in a very controversial way.
Its original structure, within a 17m long by 12m wide rectangle was very similar to what would become 
of the church of Santullano had it not had the three longitudinal naves and the two compartments at both sides of the portico. In fact, it is formed by a flat chevet with three apses, separated in the outside by buttresses, of which the central one was vaulted and the lateral ones were covered by a roof in just one plan, though its original orientation is unknown. They had a window in every apse besides another one that was found during the first excavations, formed by three horseshoe arches upon columns and capitals poorly decorated, framed in an alfiz similar to the one of San Tirso de Oviedo. It has been supposed that this window was placed in the higher plan of the main chapel, upon a high chamber with no access from the inside, of which neither any remains have rested and that it had been added during its restauration, imitating the shape of the chevets of all the churches of that group. The church consists of only one trasnverse nave of 7m wide and is very high, covered by a wooden roof and with two lateral compartments of which only remains were left of the one located in the southern side, forming an ensemble very similar to the crossing nave of Santullano, but in these cases the other three naves did not exist so it was closed with a portico instead and other two compartments at the sides in the western section, to which the door opened in the middle and the windows in the lateral compartments, that in both cases ended in round arches in brick.
Another interesting aspect is the existence 
sof an almost square construction at the southwest of the church, of which only a few walls up to 5m high had survived. They are supposed to have been the remains of a tower, rebuilt in that shape today. It would be very important if we could finally confirm that it actually was a tower and that it belongs to the same period of the church since then we would find an answer to all the doubts regarding the rests of some isolated towers built apart of the church in other Visigothic and Asturian churches, from which it has not been able to confirm that the tower belonged to the same period of the church.
In its interior, the separation between the three chapels and the nave was made with round arches in brick. The only original one left its the one at the side of the epistle as well as the column and the capital of the right side of the central one and the base of an inner door with its supporting slot. During the excavations, a flat altar table appeared with a slot in the lower part to fit it to the base and that it is now located in the southern chapel, and four fragments of an inner door, all of them decorated with vegetable motifs but of two different styles; ones are chiseled, of clear Visigothic influence and the others with more rounded shapes that already start to announce the typical Asturian style. The decoration was completed with paintings on the inner walls of which some rests have been preserved in the arch of the chapel of the epistle as well as a plinth and a modillion. They are of Roman descent and probably from the same studio than Santullano.
Unfortunately, the savagery that hovered on a great part of the Pre-Romanesque Asturian buildings between 1934 and 1936 that seriously affected Santa María de Bendones, 
prevent us from analysing with the slightest guaranty the meaning of a church that, had it survived to our days as it stood for almost eleven centuries, it would have let us know much better the changes produced in the structure of the churches along the reign of Alphonse the Chaste.
In fact, if we may consider that the original church was similar to the one rebuilt, we find ourselves in front of a structure that fits perfectly in the same group than Santullano and San Pedro de Nora; with a similar chevet in both of them; the transverse nave, very high and with side compartments of the same type of that of the crossing of Santullano, and the tripartite narthex of Nora, though with two differences that are worth to analyse: the first one would be the fact that only the central chapel was vaulted, and the second one, even more important from our point of view, is the lack of the three longitudinal naves -the ones devoted to the faithful in the Spanish rite of that time- which are basic in the rest of the Asturian Pre-Romanesque Art.
A first interpretation of these two differences would be to consider Bendones as an earlier construction to the mentioned ones, built before than Alphonse the Second 
and Tioda defined a basic type of church which at large it would be used as a base to define the prototype. This would explain why the lateral apses were still not vaulted and that the problems for the cult dedicating such a large space to the clergy and just a narthex to the faithful, originated the enlargement of the prototype structure of later built churches by adding longitudinal naves.
Another possibility which seems to us less valid, would be that Bendones was of the same period than the rest of the group, but that it was conceived as a monastic church, located in a poorly inhabited area or exclusively devoted to monks, where the model suffered some modifications to adjust it to the very different needs to those that could be present at that moment in a town more densely inhabited and therefore with greater need of space for the faithful. But this would imply some flexibility in the application of the normalized model that is not found again in the constructions of Alphonse the Second and Alphonse the Third.
Notwithstanding, no conclusion can be a definite one, since the basic problem is that we are ignorant regarding the original design of the church, as it seems there is no description left nor any graphic information about its state prior to 1936. In fact, in its reconstruction, an important part of very differentiating elements has been included, of which there was not any proof at all of their existence, and as it also happened in Santianes de Pravia according to our point of view, maybe they were exceedingly inspired in Santullano. As a matter of fact, even that the plan is known and there are not any doubts regarding that, in its development in height there are solutions that although they seem adequate to us, there is no guaranty that they existed in the original design, like the chamber created on top of the central chapel and the arrangement of the roof of the side chapels forming an ensemble of a testero similar to Nora and Santullano; the separation elements of the narthex in three compartments and even the tower; all of them included in the present structure of the church, without any proof that that was its original design but also without any prooof that it was not.
Nearest friendly lodgings of Asturian Pre-Romanesque Art
SANTA MARÍA DE BENDONES
