p. 259

            The  Buddhist  and Jaina  remains  of Dhauli  and
        Khandagiri in Orissa are well known. Both these sites
        lie from  3 to 4 miles  to the east  and west  of the
        village of Bhubaneswar. Bhubaneswar itself is full of
        temples  dating  from  the 8th century  onwards;  but
        nothing   of  an  earlier   date  had  been  hitherto
        discovered  in  or near  about  that  village.  Quite
        lately  however, a numher  of  interesting  monuments
        have  been found, some  of which  may date  from  the
        early centuries of the Christian era.

            About  three  years   ago  Professor   Radhakumud
        Mukherji  of  Lucknow  first  discovered  a piece  of
        railing  on a field lying  to the north of the temple
        of Bhaskareswar, which is locally known as Megheswar.
        It was the corner-post of a railing having lenticular
        sockets  in two  adjacent  sides  and  two  turbanned
        figures in an attitude of adoration in the other two.
        About  the  end  of  last  June,  we  made  a  second
        excursion  to that  area  under  the auspices  of the
        Orissa Historical Association and fortunately  dug up
        what  seems  to be the  rest  of the first  piece  of
        railing-post  at a spot which was about twenty  yards
        away from the place where the first had been found. A
        friend  of  mine,  Mr.  Dhirendra  Nath  Basak,  also
        discovered a third piece of railing-post with sockets
        for insertion  in opposite  sides and praying figures
        on the other two at almost  the same place.  The last
        two pieces  lay buried in the soil and only a part of
        them was exposed.

            The place where railing-posts  have been found is
        near  a  slight  mound  upon  which  the  temple   of
        Bhaskareswar  is situated.  This  temple  is of great
        interest  from  more  points  of view than  one.  The
        lingam  which is enshrined  here is of immense  size,
        being 3'--11"  in diameter  just above the Gauripatta
        and  8'--8  1/2 at a height  of about  5' above  that
        level.  The taper  of the column  is 2" in 5'.  It is
        made of a single block

                               p. 260

        of sandstone.  The shaft bears rough chisel-marks  up
        to a certain  height, above  which  the top seems  to
        have been hewn off in large fragments. It was held by
        Rajendra  Lal Mitra  that  the lingam  was really  an
        Asokan lat, which had been subsequently  defaced  and
        adapted  by Saivas  to suit their  own purpose.*  The
        column is made of fine grained sandstone and does not
        bear even the slightest  trace  of any polish  on its
        surface.  That practically  rules out the possibility
        of an Asokan origin, as all of the Emperor's  pillars
        are known  to bear a characteristically  fine polish.
        It is however  not unlikely  that "pillar was erected
        here by some king after Asoka's fashion and a railing
        was also set up all around, the remains  of which now
        lie scattered and buried in the soil at some distance
        from the mound.  The connection  between  the railing
        and the column  in the mound  seems  to be apparently
        original; the chances of the adaptation of the column
        to a lingam  being comparatively  recent is therefore
        great.  Near the temple of Rameswar, about 11/2 miles
        west  from  the  temple  of Bhaskareswar  as the crow
        files, there is a large  tank called  Asoka Kunda.  A
        friend  of  ours, Mr.  Achyuta  Kumar  Mitra  of  the
        Mayurbjanj Arehaeological  Survey, drew our attention
        to  a  most  important   object   which  lay  at  the
        north-east corner of the tank. It is the bell-capital
        of a column  of immense  size, being 4'--5" in height
        and as much  as 19'  in circumference  at its  widest
        part.  There  is a plain  surface  at the top  of the
        pillar, which  had  a raised  edge  an around  it, in
        older  to hold  securely  some  piece  of  sculpture.
        Beneath  the ribbed bell-shaped  portion  is a frieze
        adorned with scroll, flowers and animal figures.  The
        under side of the capital  has a socket for insertion
        of the shaft  of a column, the shape  of which  is in
        all probabilities hemispherical.  The diameter of the
        whole at the bottom is nearly 3'--3.8".

            We shall try to calculate  the height  of a shaft
        which might have accommodated  the bell-capital  with
        data  gathered  from columns  with  bell-capitals  in
        other parts of India.(1) The height
        ______________________

        * Mitra: Antiquities  of  Orissa, Vol  II, pp.  89-90
          Such wild theories are not seriously  regarded  at
          the present day.--Ed.

                               p. 261

        of the bell-capital  of Asoka  at Bakhira  is 2"--10"
        while the base and shaft are together nearly 38' high
        (V.A. Smith: Asoka, pages 117-118). The height of the
        column  in question  would accordingly  be about 52'.
        (2) The height of the bell-capital  of Heliodorous is
        1'--5" while the base and shaft together  are 15'--7"
        high.  If we suppose  that  the  present  column  was
        modelled  in the same  proportion, the height  of the
        base and shaft would be about 44'.

            Let  us hypothetically  assume  that  the present
        bell-capital  belonged to the column of Bhaskareswar,
        for no other column  of like size is known in or near
        about  Bhubaneswar;  and then see to what conclusions
        it carries  us.  The present  base  of the column  is
        3'--11" in diameter.  The taper is 2 1/2" in 5';  the
        column  should  therefore  have  been  15' above  the
        present  base in order  to fit into the whole  in the
        bell-capital   with  the  diameter  of  3'--3.8".   A
        bell-capital  of 4' height, and perhaps mounted  with
        an animal  figure of similar  height  resting  upon a
        column only 15' long would be absurd.  So that a part
        of the column must be embedded in the mound; and this
        portion may be anywhere from 29 to 33' in length.  On
        the other  hand we should  also keep clearly  in mind
        that the bell-capital  and this column may be totally
        unoonnected, a conclusion  which is slightly favoured
        by the petrological  character  of the two specimens.
        The  column  is made  of  fine  sandstone, while  the
        capital  seems  to be made of sandstone  of a coarser
        variety.  Two  lines  of future  research  therefore,
        present  themselves  to us.  We have  to  search  the
        jungles  of Bhubaneswar  for  a column  of the  above
        description,  and  also  to  expose  the  column   of
        Bhaskarswar  to its root, in order  to determine  its
        actual  length and see if it bore any inscription  or
        not.   This  operation  is,  however,  likely  to  be
        seriously  objected  to  by the  inhabitants  of  the
        village.

            It  is  not  possible  to say  anything  definite
        regarding  the  age  of  these  ruins.  The  youngest
        railings  that have been discovered  in India  are of
        the 3rd century A.D. But lately the Orissa Historical
        Association  has also discovered a small railing-post
        at  Lalitgiri  in Cuttack  district, the  remains  of
        which place

                               p. 262

        appear to belong to the 9th--10th.  century  A.D.  on
        epigraphic grounds.  It is thus clear that the use of
        railings  was carried  down  in Buddhist  centres  in
        Orissa  till  as  late  as  the  10th  century   A.D.
        approximately.  The very  fact that it is a piece  of
        railing does not therefore carry us far in the matter
        of dating.  Stylistically however, the figures on the
        Bhubaneswar railing-pieces are old.  The heads of the
        human figures are turbanned, with occasionally a knot
        done somewhat in the fashion of Barhut. The nature of
        the turbans, the gloved  hand, the high boots and the
        short  clothing  suggest  however, that  the  figures
        should be equated  with similar  figures  in the Rani
        Gumpha  at Udayagiri.  The modelling  is more  in the
        round, the  folded  hands  do not lie flat  upon  the
        breast  but  are  pointed  forwards.  It is  possible
        therefore that this sculpture is posterior to Barhut.
        This  is perhaps  all  that  can  be said  at present
        regarding the date of the railing-pieces.

            The date  of the bell-capital  is almost  equally
        uncertain.  From a lack of polish it is clear that it
        does not belong  to the reign of Asoka, but is likely
        to be later.  The frieze  at the base  of the capital
        bears a scroll unlike  those on Asokan  pillars.  Our
        guide, Mr.  Mitra, said that an eminent archaeologist
        to whom  he had pointed  it out, had said that it was
        similar  to a frieze on a Sunga capital.  This may or
        may  not be correct;  we should  rather  search  more
        assiduously  in the jungles of Bhubaneswar  and carry
        on excavations  in the mound  of Bhaskareswar  before
        pronouncing  anything very definite regarding the age
        of the present finds. Only this much can be said with
        certainty that these remains are far earlier than the
        oldest temples in Bhubaneswar  and that they might be
        near  the  Udayagiri  sculptures   in  the  point  of
        antiquity. They are probably later than Barhut.

            The three pieces of railing  have been lodged for
        the present at Roy's Museum in Puri.