Leaving three or four exclusions, all the outliving Gupta temples are based in Madhya Pradesh. The numerous scores remaining from the post-Gupta period, on the other hand, are dotted over an area expanding from the Indian desert in the west to Bihar in the east, between the Jamuna (Yamuna) and the Narmada, and including the present states of Gujarat, Rajasthan, Madhya Pradesh and Bihar. The maximum denseness is towards the centre, in Malwa, the prehistoric Malava. In spite of this dispersion and a groundbreaking and observational fashion, they incline to share the same themes and inspiration. The bigger temples are quite few, and not one persists in its original form; time has taken a heavy toll. Sculpture is still permeated with much of the dynamism and sensualness of Gupta times, with an appended element of baroque overstatement, while the actual material of the temple lends itself more and more to sculptural treatment. Highly observational, diverse, prosperous and even flamboyant, sometimes scattered, the temple architecture, and to a lesser amount, sculpture of the Gupta period is the most stirring in all Indian art, and in western India and Madhya Pradesh it can be still be seen in comparative large quantity.

A number of interesting and normally tiny shrines in Sauratra, in the southwest corner of Gujarat, may include the oldest surviving structures in western India-unless (as suggested by their comparative isolation) they are later than they seem, with fossilised ruins of earlier forms of superstructure nor conserved elsewhere. They are predominantly bhiimiprasadas, somewhat pyramidal, with tiers of kapotas, exemplifying storeys and a baffling assortment of other components- in the corners, small square pavilions, normally in the southern form, alternate with amalakas, and on certain occasions the superstructure is topped by a Dravida shikhara. Gavasakas, simple, unembellished, and of a size and audacity quite unknown at the time elsewhere in India, lend these temples a stark individualism, though the common effect of their plain walls, without recesses or expression and with trivial or no carving, is rather terrible. In at least two temples (Nos.2 and 5 at Bharyasara) the superstructures are more Dravidian (southern) compared to Nagara (northern), moreover, sanctums and hall are nearly enclosed by a single rectangular wall, even where there is a more conservative Nagara superstructure. Few of the more distinctive features of these shrines last or leave any eternal mark on the prevailing fashion of western India- the exception is the phiitizsanii (wedge-shaped) roof.

The temple at Gop, though partially in ruins, is by far the finest, as well as the best known of these shrines, a reminder of what must have been the architectural resplendencies of the totally gone astray Valabhi, capital of the region during the 6th, 7th and 8th centuries and historically a place of considerable significance. The temple stands on a large jagati with projections housing recesses, itself on a high base and surmounted by a double set of mouldings with a dado between them. The temple is pointed towards the east, where the platform is elongated to house two counterbalanced flights of steps each beginning- conceivably for the first time outside Andhra and Sri Lanka- with a semicircular protrusion or `moonstone` (ardhchandra). The temple itself is sandhara, i.e. with an internal circumferring passageway around the inner sanctum. There are three protrusions on a side, incorporating recesses. A frieze of garyas is another early western Chalukya feature. The entryway appears to have been unadorned. The external walls of are absent from a course above the frieze, and one can only hypothecate how the pradakshinapatha`s roof was linked to the centre, now soaring severely like a medieval European castle keep. Its inimitable and exceedingly graceful superstructure, comprising two sloping rectilinear roofs superimposed, the lower bearing two tremendous gavasakas to a side, the upper one a single one, is the earliest instance of a phamsana roof. It is surmounted by a finial shaped, like a bell (ghanta), not by an amalaka, a significant difference. The gavasakas are essentially hollow dormers in which stone images, not reliefs were placed- remarkable, because, excluding where they figure as the section of a barrel roof or a protrusion in front of the tower (iukantisa), gavasakas, which flourish on the temples of India, are usually superficial and to be measured in inches. The twice repeated rectilinear pent-roof is to be found in Kashmir temples of the 8th century and succeeding centuries (eg. Pandrethan). The phamsana roof vanishes hereafter from foremost shrines, excluding Kashmir and certain Himalayan regions, though it remains over huge halls in western India and Madhya Pradesh. Not unlike, nonetheless, are the superstructures of later mandapas in north India and Orissa and the low heights of some shrines in Karnataka, where the serried ranks of attics seem to originate from the extensively separated kapotas of the bhiimiprasada.

What slight sculptural ruins at Gop comprise a Vishnu and a Skanda, two door keepers, a Rama, and a seated woman in one of the gavasakas of the roof (a Ganapati has vanished). The style suggests a late 6th century date and, as with some of the other temples, a link with the early western Chalukyas.

A group of tiny unicellular temples spread out to the east and west of Betwa in central Madhya Pradesh can be assigned to approximately the 7th century. Many are of partially megalithic built, with enormous flat slabs for base mouldings, singular upright ones for pilasters and the intervening walls, and single slabs for the roofs. Unrefined little dolmenoid shrines can be witnessed in India wherever stone is abundant, and it has rightly been stressed that there is a conceptual relationship between them and the garbhagriha, basically a cube, its interior unilluminated, but it seems likely that some of originally had `northern` shikharas, as witness to two very alike shrines of megalithic construction (Nos. 14 & 15) standing side by side at Batesvar, one with a shikhara, the other without and thus `flat-roofed`. Erected on a monolithic slab, not attached in any way to the lower part of the shrines, the shikharas are unsteady and can easily be demolished.

These temples, surely anteceded by a simple two-columned porch, are outstanding. Though constructed of single blocks or sheets of stone, their bases, walls and pilasters are wrought and engraved in a fairly regional variant of the contemporary fashion, so as to be indistinguishable from those of ashlars masonry temples. The earliest, like Mahua Temple and the Mahadeva at Ladhaura, have columns and pilasters with episodic shafts and little corner pendentatives, sometimes plain, sometimes with large lotus medallions in the Vakataka fashion, soon to be substituted by the representative post-Gupta pillar and pilaster with a pot and foliation, both as capital and base..

The Shiva temple at Kusuma, west of Mt. Abu, known as the Ramachandraji, is still a work of enormous essence and exquisiteness, regardless of much reconstruction in current years. Presumably dated 636-7, and thus exactly contemporary with the Meguti at Aihole, it is one of only three significant temples in western India, apart from Sauragra, which definitely belong to the 7th or early 8th centuries. Within its rectangular overall plan is included a sanctum (now completely rebuilt), adequately illuminated by extended openings on the three sides of the enclosing wall, a hall with four central pillars, and one surviving ceiling over the first bay, the first of those extensively cusped and arched ceilings, which will become one of the grand glorifications of western India, and especially Gujarat. Here alone are the wooden origins of the type led astray. Except for the two questioning internal sub-shrines facing each other on each side of the mandapa, with pillared porches and elaborate split gavasaka pediments where for the first time the lotus-diamond (puparatna) and half-diamond (ardharatna) motifs appear, the plan is similar to that of the Cikki temple in Aihole. Other remarkable affinities with early western Chalukya temples of the 7th and 8th centuries comprise the central pillars with brackets supported by the protruding busts of youthful figurines .

(kumara) and the lofty figured clerestory of the fundamental nave. The entryway to the shrine, now sadly demolished and substituted but luckily recorded, was one of the finest in India, with pediment of exceptional size and conception, its adequately figured immortals in the purest and most affecting early post-Gupta style. The principal gods are Brahma, Vushnu and Lakulisa (Shiva), the first of the rows of deities exhibited on the lintels of sanctum doors of later temples. The exquisite Sadashiva is the earliest in western India.

Unluckily, only the lower portion, much of it wholly defaced, remains of the original structure of the SitaleSara temple, on the tiny Chandrabhaga, together with some sculpture, no longer in situ. The date of 689 is not unanimously accepted, but the shrine irrefutably belongs to 7th century and thus a prized document of architectural history, situated as it is near Jhalrapa and the upper tributaries of the Chambal on the boundaries of Rajasthan and western Madhya Pradesh.

In the more abundant temples of the 8th and 9th centuries, the widespread features of the post-Gupta style in western India and Madhya Pradesh emerge. The superstructure of the shrine proper almost customarily consists of a northern or Nagara tower of the simple or Latina type, i.e. without the accumulation of smaller sikharas, which is such a prevailing facet of the superior temples of the following period.

On the façade of the superstructure is a large, approximately triangular protrusion (iukaniisa niisika), sometimes bearing a recess, but always fundamentally shaped as, or commanded by, a large gavasaka. In tiny shrines, it often protrudes over the porch; in bigger ones it normally stretches over an antechamber (ardhamatJqapa or antariila). The shikhara itself is always coroneted by a huge amalaka, surmounted by a (generally missing) pot and finial. The exterior of the tower is a web of gavasakas in superimposed rows, accentuated by amalakas, set in the corners and sometimes at those of the central offset as well. All these characteristics are common everywhere, including Orissa and Karnataka. Unusual on the other hand are sanctuaries, finishing in a barrel roof, called a Valabhi in the west. The chief shrines in the bigger temples have an internal circumference, seldomly illuminated by windows, but habitually with open balconies. Shrines grouped in a quincunx (panciiyatana) occur, each continued by a tiny porch. In the bigger temples, more recurrently in the western part of the region, mandapas, both closed and open, emerge between the primary shrine and porch, their roofs either of the phamsana or of the samvaratii kind, another low domical shape, with simhakatJas (large triangular pediments ordinarily composed of gavaksas) on three sides above the cornice.

The bases, soon expanded, are of a design, observed all over the country. A vertical component, rounded at the top (kamahi or khura), takes the place of the right-angled pedestal (jagati) of south India. Above are two or more mouldings, one unquestionably a kapota; in between, serried rows of ornamented square bosses, formerly rafter ends, ordinarily replace the omnipresent kumuda (semi-circular moulding), here named a kalaia (stone jar), round or octagonal in section, of south India. In all, leaving the smallest and simplest shrines, the wall surfaces are broken up into and by perpendicular elements, chiefly the protrusions (bhadras), which carry all the way up the tower to the seat of the top amalaka. The shrine with a single cardinal protrusion per wall is known as triratha, with a supplementary offset on each side pancharatha; these are indispensable differentiations in the expansion of the temple fabric. There may be only one essential recess, or there may be one on both sides as well, or, in case of a pancharatha shrine, two. Complete sections of wall may be pilastered, but never- as normally happens in Tamil Nadu- are the protrusions, often more diverse in design and alternating with bays, cast by pilasters at the corners. The cornice consists of a kapota, frequently two. Figurines or narrative vistas in relief are nearly missing from the walls.

Image recesses and their pediments, carved ceilings (especially in the west), pillars and pilasters, all become more and more metaphorical, though the structure and the figurative implication initiated during the Gupta period are unremittingly recollected. The pillar with a pot and foliation at the base and another for capital, and the pilasters modelled upon it, is practically the style mark of post-Gupta fashion. The capital or abacus with pendant fan-palms in the corners is not a predominantly triumphant growth, because it seems too strongly modelled on the pot and foliation. Sections of pillars are often ribbed, and sub-bases appear, some with tiny recesses. Flat pilasters framing recesses often generates an innovative type of pillar, thin, round, with a serrated ring at mid-height and a little ribbed capital, and all but free-standing, sustaining as they do a little stone chadya, coffered or ribbed. The pediments evolve from extended or split gavasakas into an intricate honeycomb of tiny gavasakas, tier upon tier- one of the most distinguishing omnipresent creations of the post-Gupta era.

The post-Gupta repertory of mouldings and (especially of alternating triangles) can be seen over the whole region, though narrative friezes appear to be restricted to the west. Intertwined garlands under the cornice framing bells or half-rosettes sometimes give way to interlooped garlands. Half-rosettes rather than birds populate the gavasakas along the lower edge of the base kapota. Entryways even in very tiny shrines, grow progressively sophisticated, although fundamentally based on Gupta models and using Gupta elements, all in a single plane. At the foot on both sides stand the river-goddesses, generally escorted by attendees, with a door keeper. The entryway is flanked by superimposed panels with mithunas or additional human statuettes, normally with pilasters or both sides, and framed overall by five or six mouldings. Novel is a braided band, possibly symbolising the tails of serpents. None of these castings can compare with the laterally cut vegetal Gupta ones or the fan-palm band. The ridge, with lions and other motifs on its embossed outer edge, is normally anteceded by a moonstone. The primary innovations, however, consist of the omnipresent bimba-laliita, recurrently Garuva, a miniature figurine, half balanced from the centre of the lintel, and the common overloading at the top with immortals of all kinds, normally in recesses and pavilions. With so many specifications, the entryways often seem purely finical, but their bases can offer some of the most striking statuette groups in all Indian sculpture.

Temples in the same region, at Bhumdana, Lamba and Bhavanipura, strongly correspond to the Osian temples in fashion. Lest too much importance be laid on a provincial style, however, it is well to remember that they are not very dissimilar from the temple at Baijnath, on the longitude of Allahabad, some 54 miles (860km) to the east. Furthermore, the temples at Buchkala, comparatively nearby, more strongly resemble those at, on the precincts of Gujarat, with their squat tiny recesses with low pediments, and very few of them at that.

At Chittor (Cittau), Mewar`s ancient place forte, at least one temple, the Ksemankari, next to the Kalika Mata, is nearly an imitation of the temple at Bhavanipura near Osian, with parallel laminated pilasters instead of recesses on both sides of the fundamental bhadra. All the recesses have round ringed pillars and stone awnings over them. Similar to Osian, Chittor has two great post-Gupta Gandhara foundations- also pretty modified and added to.

A group of tiny temples at Roda in northernmost Gujarat belongs to the late 8th century. Approximately contemporary with the Osian temples, they are distinctly dissimilar in élan. The general opinion of these temples is of squatness, the shikharas low and overwhelmingly colossal, their sequential tiers undiminished by the alternation and littler gavasakas as at Osian. The web of the similar elements over the exceptional recesses is also unduly tiny and squat, as are the recesses themselves. In the identical region, the Harishchandrani Cori at Samalajihas an open balconied mandapa, with the standard half-walls and pillars, which continue towards the porch.

Most of the exquisite sculpture is perhaps a little late in date. The dressing of the hair of the fine Mother and Child (a preferred theme of the province) from Koyarka reappears in the early post-Gupta period all the way to Orissa, in Ellora and as far south as Aihole. A Saiva dvarapala, also from Samalaji, in the similar fashion is in the Prince of Wales Museum, Mumbai.

Quite a few brass images have been discovered, remarkably in Jain caches from Akota and Vasantagadh. Among the earlier ones the camaradhiirini, a female attendee, bearing a yaktail, flywhisk, a symbol of monarchs, from Akota, portrays the luxurious modelling of which the sculptor was proficient when freed from the limitation of portraying Jinas in their rigid postures.

Three comparatively large temples in Madhya Pradesh- the Gadarmal at Badoh (Pahari) near Eran and the Jaraika Mata at Barwa Nagar, both partly ruined, and the Maladevi at Gyaraspur near Vidisa, date back to the late 9th and 10th centuries. The first two possesses rectangular sanctums, but are surmounted by gigantic Nagara shikharas. Both are Gandhara. The smaller Barwa Sagar temple has a balcony on the central offset of each wall, excluding the back one, and maybe the largest and most sophisticated of all post-Gupta entryways. The Badoh temple stands on a flamboyant plinth, bordered by seven supplementary shrines. It has, in addition, an open hall with elephant protomes, protruding from the outer columns, like in Osian. The recesses are shaded with corrugated stone awnings. Both temples house tiny recesses among the base mouldings, with their own pediments, in line with the primary niches above, whose significantly lengthened gavasakas then extend upto the valance, where they are continued in a dissimilar form by one of the rat has (protruding elements) of the superstructure, thus giving the effect of a single incessant buttress from the base, or even the sub-base, upto the coronating amalaka. Both have heart- or leaf-shaped pendants beneath some of their mouldings, much bigger in Gadarmal, which is rather eccentric in the ordering of its exterior components. The Barwa Sagar temple, on the other hand, portrays an extraordinarily pure instance of the post-Gupta order, with delightfully carved beam ends in the pelmet.

At the least three post-Gupta temples in Madhya Pradesh possess triple sanctums, side by side. In the temple at Menal, possibly dating back to the late 8yh century, the shrines, though quite near together, and sharing a lengthened common porch, are divergent units from base to Nagara superstructure. Even there, however, the columns are clustered in pairs. At the backside, the two exterior shrines have a recess in the central offset, while the two auxiliary offsets are handled as ornamented pilasters; the central shrine, however, while analogous on plan at the backside, has wholly unadorned walls there, discerned only by a median band, as in the Trimurti (Temple IV) at Baroli. This serves as a beneficial reminder that differences in temple-building modes may account for what at first appear to be provincial conflicts or growths in style

Other harbingers of the next phase to be seen in temples attributed to the late 9th or early 10th centuries, include two superimposed registers of murtis, often without recesses (Kadwaha Nos.3 and 5), a plethora of vyalas and apsarases, and an inclination to initiate fresh elements, like the recesses, to the shikhara, obscuring the clear-cut boundary with the walls and giving an effect of haphazardness and perplexity (Ranmukhesvara at Kukurnath, Temple No.2 at Surwaya). Eventually, the deep but flat chiselling of such details as gavasakas, which seems to have sprung up Gujarat, had multiplied throughout the whole area.

A few dancing figurines, possibly originally in red sandstone, all severely smashed, of a couple of centuries earlier, closely resemble the dancing Indra from Kotah and the well-known Nataraja from Ujjain, both now in the Archaeological Museum, Gwalior. They possibly approximate the post-Gupta norm. Artistically incorporated, exceedingly sculptural, unmannerly though they are, the Gupta magic is missed now a days its power, its naturalism and even its idiosyncrasies.