While waiting for Robert Venturi's designs for the National Gallery extension, we have invited Mark Dziewulski to present his drawings of how the extension should work. These drawings form the core of Dziewulski's Masters thesis under the guiding eye of Michael Graves at Princeton. They are particularly interesting for showing an English architect's response to a very English problem while researching in a distinctly American context. They also show, as Dziewulski explains below, how hindsight can help design: Dziewulski worked on stage two of ABK's submission and has learnt appropriate lessons. I can't help thinking that the building should be something like this. Dziewulski's plan is in harmony with the existing gallery, yet uses powerful devices such as the great stair turret to pivot, point and place visitors through a familiar and sensible gallery grid. In terms of the contribution the building makes to Trafalgar Square, the drawings reveal a bold building that adds to its setting without unsettling it - so the attic theme of Canada House is reworked in the new building, as is the rhythm of Wilkin's facade. There is no vacant, windy courtyard opening directly off the street -instead the building runs properly along the street, screening its interior spaces. This is a graceful Anglo - American solution. Will Venturi play his cards so nicely?

Mark Dziewulski writes: This five month' investigation into the possibilities for an extension to the National Gallery was completed in June 1985. It was carried out as a thesis for a Masters Degree in Architecture at Princeton University, with the helpful advice of the architect, Michael Graves.

My first contact with the proposed extension was in the office of Ahrends Burton and Koralek, where I worked from October 1982 to July 1983 before leaving for Princeton, and where I was personally involved with their stage two submission. My interest was raised by the scheme and I saw that the challenge of the site deserved a good deal of research and investigation; I was also disappointed with the general standard of the schemes submitted and displayed at the final exhibition.

I saw the possibility of researching this project at Princeton as an excellent opportunity: it allowed me to objectify the situation within an academic environment, unhampered by the restrictions placed on commercial architects, and free to explore the subject in an environment in which architects do mix practice with academic discussion. I was also able to use my first-hand experience with the competition combined with my hindsight knowledge of all the other entries and the ensuing discussion. The brief used was that given to me in outline by Michael Wilson at the National Gallery. Objectives: Trafalgar Square has developed its present form over many centuries under a variety of architects, all of whom had a grounding in an architectural tradition which allowed them to express the very different ideas of their day without disturbing the urban fabric of the Square as a whole; moreover they were able to add to its character. The modern architect must surely use many of the skills of planning which Smirke and Wilkins have demonstrated, a tradition of values however which many architects since the sixties have rejected.

Both the temerity and self-consciousness of contemporary architects is well expressed in many of the competition entries. Looking at the schemes one sees two opposite categories. Those which either 'intrude' and disrupt the Square, or those which withdraw from it completely by disassociating themselves physically or by hiding behind tree screens. Many of the more successful of the schemes fall into these two categories. Several place a courtyard between new and old and tend to face Pall Mall East. Others use a tower or some such large scale motif of 'modern architecture' to face the Square. I believe, however, that the British architectural profession can demonstrate its abilities in a far more positive manner.

It would seem that a bolder move is necessary - one which is aware of current architectural discussion and is also sympathetic with the form of London squares.

Facing the challenge and making a strong contribution within this planning tradition would seem a more positive attitude than that of shying away from the issues or making token 'grand' gestures which are eventually watered down stylistically to make them more palatable and which ultimately therefore become bland and faceless.

Description: There are two main sets of criteria central to this investigation. The first determines the nature of the floor plans, layout and connection with the existing building - the internal arrangement. The second set examines the building as an urban monument and as part of an existing context - the 'external aspects'. It was from the interaction of these two roles of the building that the design resulted.

Internal arrangement: The main street access is made up the steps facing the existing building and through the arch and entrance court. This court is surrounded by a colonnade along the top of which runs a viewing platform connected to a restaurant. The main entrance is directly opposite the flight of steps and its facade expresses its function as both portal and symbol of the museum.

The lobby of the museum is a large circular room running the full height of the building containing vertical circulation and provides a link between the public facilities on all floors. Acting also as a visual connection between the levels, it has a scale and hierarchy appropriate to its function within the building and provides an interesting ascent to the galleries themselves, where it adjoins a seating area overlooking Trafalgar Square.

The most logical extension to the original floor plan is made by continuing its main central axis east to west. The new gallery is at the same level' as the old museum, the transition between the two being made through a portal into a vestibule and its plan derives from a study of some of the more successful museums in Europe - the Uffizi, the Vatican, the Louvre and others - its basic layout is not dissimilar from Wilkin's plan. It is essentially cruciform in plan, the main entrance lobby connecting to the lower end of a long exhibition space - a tall barrel-vaulted gallery, 46 metres in length- off which there are exhibition rooms to each side. This long gallery terminates in a large room with a niche, an excellent location for the larger altar pieces, and where it crosses the east-west axis from the existing museum it opens up to become a large centrally planned space lit from a circular roof light. The main east-west axis is terminated in a niched room which acts almost as the transept in a church plan.

This plan has a number of advantages. It provides an easily remembered mental picture, since one is always aware of one's location in relation to the long gallery, and it facilitates an easy progression through the museum without the need to retrace steps, allowing the paintings to be grouped in a chronological sequence. Within this system the plan also allows a variety of room types and proportions suitable for the different sizes and styles of painting. The hierarchy of spaces, the sequence of rooms and their variety provides a flexibility in the arrangement of paintings to allow for future additions.

The display of the individual paintings is seen as a central issue in the design. Technical requirements require overhead natural lighting throughout, supplemented if necessary by diffused artificial lighting. Environmental control is critical and a large plantroom and generous space for servicing ducts is provided in the basement. The form and character of the rooms follows the needs of the collection. The planned sequence of rooms contains a hierarchy of spaces which are individually appropriate in scale for the specific groups of paintings and also sympathetic to their character. The interiors are modern yet also echo our memory of what a museum should be, while still capturing something of the original settings for which the paintings were intended. Many of the rooms have niches and are intended to display free standing paintings.

The main staircase and lobby connect the other public facilities. There is a lecture theatre to seat 500 people leading from the lower lobby, temporary exhibition space and an audio-visual room at ground floor level, each capable of independent operation when the gallery level is closed, and a large public information centre close to the main entrance, which would contain computer terminals and other facilities dealing with the contents of the museum. All these have their own attendant storage and service areas. The lower ground floor and basement contain large storage areas for the museum, publishing offices, museum bookshop, and small shops, as well as the extensive plantroom required. The storage areas are all fed by a service tunnel and lift from the main unloading area on St. Martins Street. Beneath the colonnaded plinth at the main entrance lie extensive warder facilities, and cleaning and storage areas.

At the top of the main stair and lobby sits a restaurant to seat three hundred, overlooking the Square, and giving access to the high level promenade or viewing platforms. The kitchens are adjacent and the main storage facilities are in the basement level.

The building as urban form: The building must represent its roles as both a new structure containing one of the best collections of Renaissance art in the world, and as an extension to the existing museum whose architectural character is already an important image of our national art heritage. The massing of the building within its context is also of fundamental importance. Trafalgar Square is formed by many varied buildings constructed at different times which all, however, form a sufficiently uniform frontage facing on to the Square so as to enclose it, surrounding Nelson's Column and containing the space; the National Gallery can be seen as the northern portion of this 'wall'.

The Square is not, however, symmetrical about the Gallery, the eastern arm containing St. Martin-in-the-Fields breaks back to allow St. Martins Place to enter the Square, whereas the western edge formed by Canada House runs up to a corner of the Gallery, Smirke's facade effectively facing the Hampton site. One should not, therefore, make the mistake of seeing the Square as a piece of highly symmetrical Baroque planning. It is, however, a good example of how a variety of building types and styles may combine to create a balanced and harmonious composition.

The massing of the proposed scheme takes advantage of this asymmetrical nature of the Square. By continuing the line of Canada House it effectively completes the Square to make a corner. The new colonnade faces that of Canada House and the two form a gateway into the Square. This gateway frames a view of the main portico and dome of the National Gallery from an approach along Pall Mall East. Part of the new facade follows the line of Pall Mall East, identifying the new building with that street and making the entrance to the Square more pronounced.

One of the main criticisms of the original Wilkins' Gallery has been the elongated proportions of its facades and the weakness of its unresolved ends. While the height and solidity of the new building are in keeping with the original, it is a strong enough element both to complete the edge of the Square and help anchor the corner of the old Gallery.

The requirements of keeping pedestrian access through the site seem to have been one of the most challenging issues. Rather than allowing the route to become a long, low and dark passageway beneath the new building, as proposed in many of the competition schemes, I decided to use the precedent of the may fine old shopping arcades which exist in London and in that area in particular. Research revealed that these arcades have been the traditional way to allow access through a building block and have often been associated with theatres and museums. The Royal Academy and Burlington Arcade, the now demolished Royal Opera and its arcade, which still exists, are good examples and many others, of which Lowther Arcade is probably the closest. These have lasted through the years because of their practicality as well as popularity, and are elegant and convenient pedestrian thorough-fares.

The advantages of using an arcade seemed enormous. The passage through the site would become a positive asset, a pleasant journey, and should make it a popular route connecting Leicester and Trafalgar Squares. It also makes it possible to give the new museum bookshop an extensive street level frontage without disrupting the exterior of the museum -the street facades. The problem of the growing need to provide street level retail space in new museums has become common (as I myself have experienced when working on the Whitney Museum in New York, in the office of Michael Graves). An arcade also allows natural light to enter the lower floors within the building which contain offices. Several shops - rare bookshops, etc. - can be accommodated, bringing in revenue whilst again enhancing the route through the site. The space would afford an exciting transition through the museum itself, across a bridge which looks into the arcade.

The geometry of plan of the building is simple. The main entrance courtyard, its colonnade and the shopping arcade all sit on a long axis which, starts in the centre of Canada House's 'temple-front' facade and runs all the way up St. Martins Street to Leicester Square. The tracing of this axis falls into place with surprising ease, the front of Canada House facing as it does the axis from Leicester Square.

The entrance to the arcade on St. Martins Street also allows a gesture of significant scale towards Leicester Square from where it can be seen framed by the buildings on either side of the street. The entrance to Trafalgar Square would also be an appropriately interesting one down the main steps beneath the archway, facing the main building's entrance.

The character of the façades is seen to represent both the role of the building as a contemporary structure and its role as a public image for a new museum of one of the most important collections of Renaissance painting in the world. The building's materials are both modern and traditional - steel, glass and stone - and display a narrative of construction using beam, column and truss, combined however in a way sympathetic with the classical proportions and precedent of the existing building. It picks up many of the decorative and symbolic elements of the original building and interprets them in a modern language.

The building is modern and speaks of our present culture; it is neither historical, pastiche nor futuristic. We should not be seeking for replicas of past styles that have meaning for us only by romantic association and that deny the effect that the modern movement has had on our culture.

Also, as many buildings have already demonstrated, and many of the competition schemes themselves, what may seem futuristic today, soon becomes the shallow and theatrical of tomorrow. One must ask oneself for how long a building is considered new? Ten, twenty or even fifty years are all very short periods of time in the life of a building. We should not be searching for the 'StarTrek' of tomorrow but instead for something timeless, something with lasting value and meaning, which reflects our present understanding of art, culture and urban fabric.